April 27-30, 2022; Barcelona, Spain (+Virtual); Full Report – Draft

The ATTD organizers once again blew us away with the standout event that was the 15^th annual ATTD conference, back in person in Barcelona after a virtual conference in 2021. Immediately below, you’ll find our top themes from the meeting, followed by highlights in the categories below. New highlights (those that weren’t included in our day-of reports) are highlighted in blue. Note that some talks may appear in multiple sections.

• Glucose Monitoring
• Insulin Delivery
• Digital Health, Decision Support, and Telehealth
• Time in Range and Beyond A1c
• Diabetes Therapies
• Big Picture
• Posters
• ATTD Yearbook
• Exhibit Hall

Themes

1. Double victory in Barcelona and at home: ATTD 2022 delivers successful virtual and in-person experience as the first major hybrid conference

We were blown away by how smoothly ATTD 2022 was run and we applaud Drs. Moshe Phillips and Tadej Battelino for pulling it off! We were treated to a top-notch experience in Barcelona, where conference halls, meeting rooms, and post-conference dinners were abuzz with discussion of the top learnings in diabetes technology. The exhibit hall captivated our team with its elaborate booths, helpful demonstrations, and interactive displays, not to mention Abbott's espresso bar and Novo Nordisk's relaxation lounge. For virtual attendees, conference livestreaming operated without a hitch; plus, on-demand coverage allowed us to replay our favorite presentations and soak in further knowledge after the conclusion of the conference. We also appreciated the ability to check out e-posters, peruse the ATTD 2021 yearbook, appreciate the extensive attendance list, and connect with other attendees on the ATTD 2022 Virtual Platform. Those of us who attended in-person were quite impressed by the conference organizer’s attention to COVID-19 precautions, including offering rapid testing at the conference center. Beyond conference sessions, our team also enjoyed the numerous social opportunities, including the ATTD Run to Overcome Clinical Inertia along Barcelona’s beautiful beaches and the awe-inspiring flamenco performance at the Opening Ceremony. Ultimately, we were impressed by the extremely high quality of conference sessions and high engagement among attendees, faculty, people with diabetes, and industry representatives, and we’re thrilled that so many people were able to participate in the format of their choosing – bravo!

2. Technology takes on type 2: Continued evidence in support of CGM among type 2s both on and not on insulin; first studies evaluating AID in type 2s show promising results

ATTD 2022 continued the momentum from last year’s meeting with a focus on technology use among people with type 2 diabetes. Conference organizers Drs. Battelino and Phillip discussed their intentional emphasis on tech among type 2s during their inaugural ATTD press conference highlighting applications of CGM to help inform clinical adjustments as well as the ability of decision support software to support providers, combat clinical inertia, and drive improved outcomes for people with type 2. Across the meeting we noted three distinct areas of technology use among people with type 2 diabetes: (i) CGM use among type 2s on both insulin or non-insulin therapy; (ii) AID use among type 2s previously on MDI; and (iii) digital health and professional CGM use among type 2s. Across all three areas, data from people with type 2 diabetes who used technology was very promising and continued to demonstrate that technology has a variety of applications that expand beyond people with type 1 or those on intensive insulin therapy and can also be cost-effective. However, while technology use has been associated with improved outcomes for people with type 2 diabetes, there are still challenges in increasing adoption, which Dr. Rich Bergenstal (International Diabetes Center) addressed in a Saturday morning session on his novel C²GM algorithm for using CGM and Time in Range information, with a specific focus on Time in Range and Time Below Range, to inform clinical care for people with type 2 diabetes on insulin. For us, this session was one of the top highlights of the entire meeting and we cannot commend Dr. Bergenstal enough for his continued efforts to make CGM data actionable for providers, especially those in primary care who may be less familiar with the technology. We see tools like the C²GM algorithm with its simple and specific clinical adjustments as a key step in making providers more comfortable using CGM among their patients and also ensuring that CGM is used to its fullest potential to help drive clinical interventions alongside lifestyle and behavior change among people with type 2 diabetes.

• CGM use among type 2s on insulin continues to be associated with improvements in glycemic management and patient reported outcomes. Following last year’s landmark read-out of the MOBILE study of Dexcom G6 use among people with type 2 diabetes on basal-only treatment, this year we saw further analysis demonstrating and improvement in patient reported outcomes among people with type 2 diabetes in the MOBILE study. Data from real-world Dexcom G6 users participating in the Type 2 Help study (NCT04503239), demonstrated similar improvements in PROs among people with type 2 diabetes or prediabetes – this data is very important to demonstrate that technology can do more than drive glycemic improvements, but can also help reduce the burden of diabetes management. Furthermore, data from two large cohort studies, one out of the International Diabetes Center (n=2,331) and one from real-world Dexcom G6 users (n=69,365) demonstrated improvements in glycemic management following CGM initiation, especially among patients who engaged with their CGM data. Additionally, we learned from Dr. Battelino that there is a new consensus report in the works that will provide recommendations for the use of CGM to drive therapy intensification in people with type 2 diabetes that is expected to be published in February 2023. Following the massively influential consensus on Time in Range targets published back in 2019, we are looking forward to reading this new consensus statement next year and hope that it can drive similar change in the way type 2 diabetes in managed to help patients achieve improved outcomes and quality of life.
• ATTD 2022 was the first conference where we saw the impact of AID use in people with type 2 diabetes. While AID has been discussed as a game changer in type 1 diabetes management, its implementation for patients with type 2 has been much quieter, likely given that the majority of people with type 2 diabetes do not require basal-bolus insulin therapy. However, for those that do, we were extremely excited to see positive results from the feasibility trial of Insulet’s Omnipod 5 AID system which demonstrated a strong 3.4 hour/day improvement in Time in Range to 61% for prior MDI users and a very impressive 5.9 hour/day improvement in Time in Range to 57% for prior basal-only patients. We are extremely encouraged by these outcomes and are looking forward to seeing additional results presented at ADA 2022. Additionally, while we didn’t see any data, we learned from Dr. Roman Hovorka that researchers at the University of Cambridge have completed the analysis of an RCT evaluating a fully closed loop CamAPS system in a broad population of people with type 2 diabetes on basal-bolus therapy with plans to share this data at EASD 2022.
• Professional CGM in type 2s has also been associated with improvements in glycemic management providing an alternative method of data collection for patients who may not be able to or may not always want to use a CGM device. Specifically, data from Dexcom G6 Pro use among adults with type 2 diabetes (n=700+) demonstrated strong average A1c reductions of 0.5% for non-insulin users. We continue to think that Professional CGM is an incredible tool with the potential to provide large amounts of data to inform clinical decision-making that can be especially useful at times of transition in diabetes management. Additionally, for patients who don’t use insulin, but could benefit from learning more about how lifestyle and other medications impact their glycemic management, we see Professional CGM as a very cost-effective way to collect and interpret this information. In addition to Professional CGM, digital health continues to make inroads in the management of type 2 diabetes and we saw data from Welldoc patients who used Dexcom G6 intermittently (n=55) and experienced a remarkable 4.8 hour/day increase in Time in Range. While digital health remains an important area of diabetes management, there appear to be less understanding on how the field moves from broad excitement that surrounded major acquisitions like Teledoc/Livongo, for example. While it’s very clear that major impact is happening at multiple organizations, particularly those that are well-funded, that is certainly not true at all (e.g., Noom as one recent notable example). We look forward to talking to more experts to better understand how successes can be scaled and greater impact can be seen. As digital health was not a major topic of ATTD 2022, we look forward to looking at previous reports to better understand what data has been shown, what consensus has developed, and what thinking is among practitioners and experts there is on how to look at data on these solutions throughout the year. There is certainly a lot of potential and we hope to work on how we can focus internally to better bring these learnings to Closer Look readers – we welcome feedback on this front.

3. It’s Thrive Time: Real-world AID users continue to achieve impressive and sustained glycemic outcomes, regardless of system choice

At last year’s ATTD, we said that 2021 was the “year of real-world AID data,” but make no mistake – ATTD 2022’s lineup did not disappoint on the real-world evidence front. From our view, it’s not surprising that the bounty of real-world evidence for AID continues to expand with more impressive outcomes each year, given the continued uptake of diabetes technology around the globe, and the continued innovation driving the field forward. In 2021 alone, insulin pump sales totaled a record $3.4 billion (+18% YOY), and we estimate that well over 600,000 people rely on AID systems as their primary mode of insulin delivery. While of course there remain notable feature differences between commercial automated insulin delivery system, renowned AID-expert Dr. Boris Kovatchev so eloquently remarked during a plenary session that most (if not all) AID users routinely achieve a Time in Range in the mid-70% range, making the handful of commercial systems not so different after all. Indeed, longitudinal real-world data from the poster hall of MiniMed 780G (n=18), Control-IQ (n=18), and Diabeloop DBLG-1 (n=54) users shows that while 780G might have conferred the best GMI and relative Time in Range improvement (by a thin margin), there were no between-group significant differences in A1c at three months. Plus, we can’t forget Ms. Dana Lewis’s (Founder, #OpenAPS) captivating presentation, where she remarked that the similarities between commercial AID systems and DIY systems “outweigh [the] differences.”

• While we saw real-world outcomes from many different AID systems, it’s hard to deny that Tandem’s real-world Control-IQ data stole the show. Dr. Jordan Pinsker (Tandem) presented a first look at nine-month results from the CLIO trial, from which we’ve seen plenty of other data at past gatherings: (i) three-month data from the CLIO study in adults at ATTD 2021; (ii) a sub-analysis by racial and ethnic groups at ADA 2021; (iii) three-month quality of life data at Keystone 2021; and (iv) three-month pediatric data at ISPAD 2021. In the Control-IQ cohort, users who switched directly from MDI achieved a particularly impressive median A1c reduction from 7.9% to 7.1% after nine months (n=426). Notably, Control-IQ was shown to be effective at improving A1c in every age group and baseline insulin delivery method. Also from the CLIO study, Dr. Harsimran Singh (Tandem) presented six-month patient-reported outcomes (PROs), showing that participants reported a significant 33% reduction in the impact of diabetes on their overall wellbeing from a score of 4.66 to 3.12 (out of 10, p<0.001); in terms of device satisfaction, participants reported a 23% increase in device satisfaction from a score of 7.2 to 8.8 (p<0.001). Taken together, these data show that participants reported a significant reduction in the burden associated with diabetes management. Lastly, Barbara Davis Center’s Ms. Cari Berget presented real-world data from a prospective, observational study evaluating Control-IQ in youth with type 1 diabetes (n=183) over one year, showing that while those with A1cs >9% maintain their significant benefit over a year, those with lower A1cs don’t always consistently benefit. Unsurprisingly, the greatest benefit of Control-IQ was seen in the group with a baseline A1c >9%, in which participants saw a 3.4 hour/day Time in Range improvement from 39% at baseline to 53% at three months (maintained out to a year), and a huge 1.5% A1c improvement from 9.8% at baseline to 8.3% after three months of Control-IQ wear. The group with baseline A1c values of 7%-8.9% offered saw clinical improvements at three months, but these were not entirely maintained out to one year.
• Medtronic’s plethora of MiniMed 780G real-world outcomes were a strong continuation of the company’s data readout from EASD 2021. Dr. Robert Vigersky (Medtronic Diabetes) presented three illuminating real-world analyses from a cohort from MiniMed 780G users, in which the entire population of MiniMed 780G users in the study (n=25,396) achieved a 74% Time in Range and 6.8% GMI. Dr. Vigersky presented two other studies from this cohort: (i) a real-world crossover study (n=6,299), demonstrating that users who initiate MiniMed 780G witness a +2.7 hour/day improvement in Time in Range (+11%); and (ii) a longitudinal cohort (n=9,119) showing that the glycemic improvements are observed in the first month after initiating MiniMed 780G and sustained over six months. But who could forget – our Associates’ favorite part of the ATTD 2022 exhibit hall was the Medtronic booth, where the company shared remarkable real-world country-level Time in Range data from MiniMed 770G and MiniMed 780G users around the world. Set up as an interactive display, attendees could choose countries where Medtronic’s AID systems are available and then see average Time in Range across users in the chosen geography. Encouragingly, population-level data from both MiniMed 770G and 780G showed that, on average, users are achieving >70% Time in Range on a MiniMed AID system. In the US, data from 33,713 users who spent an average of 94% of time in SmartGuard demonstrated an average Time in Range of 71%, Time above Range of 27%, and Time Below Range of 2.9%.
• Diabeloop headlined ATTD’s Product Theater circuit with three-month real-world data, adding to our first look at data from the Diabeloop DBLG1/Roche Accu-Chek Insight/Dexcom G6 AID system at EASD 2021. CEO Mr. Erik Huneker shared real-world data from 1,914 of the first DBLG-1 users in Germany with mean Time in Range coming in at 73%. In a smaller subset (n=974) of participants between September 2021 and December 2021, Time in Range improved a very impressive +4.4 hours/day following initiation of AID (55% to 73%), with low rates of hypoglycemia. These data were also included in the poster hall, showing the Time in Range improvement achieved by DBLG1 users in Germany has steadily increased over time.
• Outside the commercial realm, it seems real-world DIY AID users are also thriving. Dr. Thomas Crabtree (University Hospitals of Derby and Burton, UK) shared initial data from the Association of British Clinical Diabetologist’s (ABCD) DIYAPS audit program, which aims to capture clinician validated data from DIY AID system users in the real world. The preliminary data presented by Dr. Crabtree include 101 people in the UK on DIY systems who averaged 41-years-old, were 90% White, and had a baseline A1c of 7.0%. Dr. Crabtree drew attention to the fact that on average, they already were achieving A1c consensus targets and that they had a long duration of diabetes (26 years). Overall, these DIY AID users saw a significant improvement in Time in Range (+4.6 hours/day; p=0.046) and A1c (-0.6%; p<0.0001), as well as a trend toward reduced time in hypoglycemia and reduced hospital admissions (p=0.063). We also learned from survey data (n=662) in the poster hall that type 1s in the US (44%), Canada (15%), and UK (11%) chose to use a DIY system for various reasons: (i) transparency (33%); (ii) interoperability (29%); (iii) support for open-source software (25%); (iv) only AID available where live (22%); (v) belief in greater safety (19%); and (vi) belief that is more affordable (12%).

4. Brilliant Biosensors! Next-gen CGMs do not disappoint with impressive accuracy and outcomes; integrations lay the framework for an interoperable future

Next-generation CGM systems took center stage at ATTD 2022, with several impressive data readouts from Dexcom G7 and FreeStyle Libre 3. We aren’t necessarily surprised that the brainpower at Abbott and Dexcom continue to iterate on past developments and create smaller, more accurate sensors with longer wear times. However, we can’t help but convey our amazement at the performance of these next-generation systems, which enable people with diabetes to understand better how lifestyle choices and treatment adjustments affect their blood glucose. There was only one logical conclusion amidst the game-changing data we saw at ATTD: CGM innovation is far from dead.

• For the first time ever (!), we saw prospective clinical accuracy data on FreeStyle Libre 3 (n=100), which was read out by Dr. Emma Wilmot (University Hospitals of Derby and Burton) during an Abbott symposium. In the study, Libre 3 had a topline MARD of 7.8%, and 93% of matched CGM-YSI pairs met the ±20/20% overall agreement rate, placing the FreeStyle Libre 3 sensor accuracy right around that of Dexcom G7 (±20/20% agreement rate of 95% and an overall MARD of 8.2% for upper arm wear and a ±20/20% agreement rate of 93% and an overall MARD of 9.1% for abdomen wear). It was good to hear that before reading out data from the trial, Dr. Wilmot confirmed that this study, conducted across four centers in the US, was used to support Abbott’s FDA submission of FreeStyle Libre 3 as an iCGM in 2021. We imagine as FDA moves ahead, there are many ways of doing this. Since Dr. Lawton confirmed that it was used to support but didn’t explicitly say more, we assume these data represented part of the what FDA received though wasn’t necessarily pivotal data. We’ll be excited to hear more about the FreeStyle Libre 3 submission package over time. While based on the study design, it seems that the data may not have come from the US pivotal trial, given that the study only included 6,845 matched CGM-YSI pairs, which would likely not be enough data to obtain iCGM designation in the eyes of FDA, we also aren’t sure that iCGM would be the goal for this submission. Additionally, in FreeStyle Libre 3’s online user manual, which emerged earlier this year online, the company reports that in a separate study with 23,503 matched pairs, the system had a MARD of 8.9%, which still definitely meets the FDA’s special controls for iCGM designation, but is not the same as the accuracy data presented during the conference. From our view, the accuracy data continues to be a marketing point, though from multiple patient perspectives (likely excluding pregnancy), it is not a big deal given that things like warm-up time, sensor size, length of wear, cost, etc. are still areas of differentiation. As background on the competition, Dexcom G7’s pivotal trial included 77,774 matched CGM and YSI data pairs – whether or not the 6,845 matched pairs in Abbott’s accuracy study are sufficient to support an iCGM FDA submission is one question – the ultimate approval is a bigger deal from our view, though AID is gaining in importance to address at some stage, given that reducing the cost of AID is a topic of very high demand.  
• We also saw safety and accuracy data from Dexcom G7 in children, which was likely used to support a pediatric type 1 FDA submission. During a highly attended Dexcom symposium, Joslin’s Dr. Lori Laffel presented results from a prospective, multicenter, single-arm Dexcom G7 safety and accuracy study in children (ages 2-17) (n=164 children with 240 sensors), demonstrating a topline MARD of 8.3% and 9.0% for arm and abdomen wear, respectively. Broken down by age group, Dexcom G7 had a MARD of 8.3% for arm wear and 9.0% for abdomen wear in children aged 7-17, and an overall MARD of 9.3% for children aged 2-6. These values are quite strong – well within the FDA special controls necessary for iCGM clearance, and strong enough to confer an advantage among other FDA cleared CGMs for children and very young children. Although Dr. Laffel did not directly refer to this study as a pediatric pivotal trial for Dexcom G7, we hope that these data could be used to support an FDA indication for pediatric type 1s, given the study’s similar design to the adult pivotal trial and the robust number of matched CGM-YSI pairs.
• Beyond these accuracy studies, a new wave of CGM integrations saw progress. On Day #1 of ATTD 2022, we learned that Abbott has forged a new partnership with Ypsomed and CamDiab to integrate FreeStyle Libre 3 with the myLife CamAPS FX AID system that was first announced in March 2022, which, when launched, would mark the first commercial AID system involving a FreeStyle Libre CGM. We also saw Dexcom G6 being used in a new context as a part of the Beta Bionics’ insulin-only iLet system, from which we saw exciting pivotal trial results on Day #4. We look forward to seeing an expanded network of CGM integrations that is powered by continued sensor innovation and industry partnerships.
• While we didn’t see any accuracy data on Dexcom’s newest ONE CGM system, we did see outcomes from real-world users (n=1,859), who spent +1.6 hours/day in Range (66% vs. 59%) after the first ten days of sensor use. Dexcom ONE maintains the same form factor as Dexcom G6 but is built on a novel and “simplified” software with a “simplified alarm scheme” (i.e., no predictive alarms). The system is not compatible with AID systems nor does it support remote monitoring or data-sharing functions. Dexcom ONE has been available in Bulgaria, Estonia, Latvia, and Lithuania since September 2021 via Dexcom’s e-commerce platform, is being launched in Spain, and will be launched in the UK. Dr. Žydrūnė Visockienė (Vilnius University, Lithuania), who presented the real-world data, noted that Dexcom should allow ONE users to use the Dexcom Follow (iOS, Android) app, which enables CGM data sharing with friends and family. On this, we sincerely agree; at the same time, it seems likely that not having predictive alerts on the Dexcom One is part of the strategy. For those who are asserting that they think that Dexcom should enable predictive alerts on its ONE CGM, we would merely note that we are sure this has been something Dexcom has considered and that it isn’t accidental that they are not included. While it’s been suggested that Dexcom’s rationale for excluding certain features from its ONE system was an attempt to ensure that the system is as simple to use as possible, it’s not our sense that that is what they are saying. If asked, we imagine Dexcom would acknowledge that features and pricing those features as part of the most advanced CGM systems was the strategy. Since the Dexcom Follow app and predictive alerts on Dexcom G6 and G7 are clearly important tools for caregivers and loved ones to help care for people with diabetes and detect emergencies such as hypoglycemia and DKA, we hope this is considered by the company.
• While the field already got a close look at accuracy data from Medtronic’s Guardian 4 Sensor at ATTD 2021, we did see pivotal data affirming the sensor’s performance as a part of MiniMed 780G. Medtronic presented poster data (EP018) demonstrating that participants in the MiniMed 780G pivotal trial who continued to the extension phase and transitioned to the Guardian 4 CGM from Guardian Sensor 3 maintained their glycemic outcomes. Medtronic showed that children under 18 (n=109) and adults ≥18 (n=67) using MiniMed 780G with Guardian 4 overall maintained an average Time in Range of 73% after three months (children: 72%; adults: 77%). Compared to the overall 75% Time in Range at the end of the MiniMed 780G pivotal trial, this poster shows that strong glycemic outcomes were sustained at three months across the entire study sample, confirming the strong accuracy of the MiniMed 780G AID system with Medtronic’s next generation Guardian 4 CGM. 

5. It’s about Time (in Range): Time in Range discussions move beyond the traditional validation framework with expanding focus on maximizing CGM metrics’ benefit in clinical practice and as a measurement tool in studies, all framed by the consensus meeting on CGM outcomes in non-device trials

Time in Range and CGM metrics were once again a hot topic of this year’s ATTD sessions, although the framework in which they were primarily discussed went beyond the traditional “Beyond A1c” framework. Rather than focusing on the validation of Time in Range (although we certainly saw some data on that as well), these discussions and presentations focused on the wide-ranging value of Time in Range in clinical care for various populations and in studies evaluating non-device interventions. To us, this demonstrates the evolution of the “Beyond A1c” movement and the progress that we’ve collectively made in expanding the acceptance of Time in Range and other CGM metrics as clinically meaningful outcomes.

• The day before ATTD 2022 began, the consensus meeting for the standardization of CGM metrics in evaluating non-device interventions convened to establish standardized recommendations for the use of CGM in non-device trials, including those that evaluate therapeutic, behavioral, and surgical interventions. Importantly, this means that the recommendations are intended to inform the use of CGM as a measurement tool rather than an intervention tool in clinical trials. The group, which is led by Dr. Tadej Battelino (University Medical Center Ljubljana, Slovenia), intends the recommendations to be used by both the FDA in evaluating trials that use CGM as a measurement tool, and by those who are running diabetes trials to better use CGM as a measurement tool. During the pre-ATTD meeting, the conversation primarily centered on the input of researchers, particularly very well-known clinical trialists, as well as experts in hypoglycemia and Time in Range, whose opinions will make up the consensus report. Additionally, industry members and the FDA’s Dr. Kristen Pluchino were present and offered interspersed commentary on the viewpoints of these other important stakeholder groups.  While the specifics of the group’s recommendations cannot yet be made public, Dr. Battelino did share several takeaways in his Day #1 talk reviewing the conversation, during which he stated that the meeting effectively updated a draft of the consensus report and importantly, added recommendations on how to statistically present CGM data and what a noninferiority margin should be for various CGM metrics in clinical trials. We see these recommendations as important in the effort toward enabling both the use of CGM metrics in clinical trials as well as to be included on the prescribing information. Having now seen what clinical trial data looks like from the InRange trial, we believe this information will be profoundly useful not only to policymakers who are assessing therapeutic submissions, but it will also be very useful for clinicians in the field who are considering what therapies are best for people with diabetes. We would like to see CGM considered for all trials, at least intermittent use, so that the value of the therapy will be easier to understand at multiple levels:
  
  • Policymaker level (FDA in particular);
  • Payer level (especially Medicaid and Medicare as well as commercial payers)
  • Clinicians; and
  • Patients.

• Also on the CGM metrics in clinical trials front, ATTD 2022 Day #3 was headlined by the readout of the Sanofi-sponsored InRange Study, the first-ever RCT to use Time in Range as a primary endpoint in comparing two second-gen basal insulin analogs. Specifically, InRange compared Toujeo (insulin glargine U300) and Tresiba (insulin degludec U100) over twelve weeks and showed that overall, there were no significant differences between the second-gen basal insulins’ Time in Range values at week twelve (nor significant differences in hypoglycemia or A1c). Beyond the actual outcomes of the trial (which are themselves meaningful), the study offered incredible learnings on the use of Time in Range as a clinical endpoint and primary outcome. It was a particularly timely readout (just three days after the consensus meeting) and offered a real-world example of how CGMs might be used as measurement tools in clinical trials moving forward.
• While less of a focus at ATTD 2022, this year’s sessions did include discussion of A1c vs. Time in Range and new data on the validation of Time in Range as a clinically meaningful outcome. In one of ATTD 2022’s final sessions, Dr. Irl Hirsch (University of Washington) championed Time in Range as an individual-level clinical outcome, noting that A1c is “critical” for population insights but can be “problematic” and “dangerous” at the individual level, an evolution from the argument that he made at Keystone 2021. Elsewhere, Dr. Jolien De Meulemeester (KU Leuven, Belgium) shared a retrospective, cross-sectional, real-world study that found that Time in Range was inversely associated with the prevalence of microvascular complications, but that the relationship between Time in Range and macrovascular outcomes was mixed, based on data from the RESCUE and FUTURE studies. The relationship between Time in Range and macrovascular outcomes certainly requires more time – while will certainly be important evidence to further grow acceptance of Time in Range, this hasn’t yet been validated in ways that it could be in the future as has been seen in DCCT, UKPDS, etc. What the world of real-world data and acceptance will be is a key area to watch for.
  • In the poster hall, a series of posters correlated CGM outcomes with a gamut of important outcomes, including the correlation between Time in Range and mobility metrics among older adults with type 2, between Time in Range and A1c based on real-world data, between glycemic variability and hypoglycemia risk, between Time in Range and sleep, and between Time in Range and retinopathy. Finally, in a pregnancy-focused session, Prof. Eleanor Scott (University of Leeds, UK) discussed the results of an analysis comparing CGM metrics over the course of pregnancy for pregnant people who had large for gestational age (LGA) babies vs. those whose babies were not LGA. The analysis showed that slightly higher mean glucose values over pregnancy and substantially higher glucose values between weeks 20-30 were correlated with LGA, a pattern that traditional metrics like A1c completely obscure. This once again illustrated the incredible value of CGM metrics in understanding – and thereby mitigating – the risks of adverse outcomes whether that be poor sleep, type 2 complications, or fetal and maternal outcomes.

6. AID marches forward: Strong results across varied populations including pediatrics and prior MDI users

AID systems are gaining steam and demonstrating strong outcomes across a variety of populations including pediatric patients with diabetes as well as adults transitioning directly from MDI. Especially among pediatric patients, we are so encouraged to see strong Time in Range outcomes that have the potential to reduce diabetes burden and help children with diabetes have greater freedom to enjoy activities that otherwise may have been challenging to balance with managing their diabetes. At ATTD 2022 specifically, we saw new data out of a pediatric diabetes camp in Italy that demonstrated improvements in Time in Range for children who started using Control-IQ. In an oral session on Control-IQ use among pediatric patients with type 1 diabetes, we were also encouraged to learn that children with higher baseline A1cs (>9%) saw substantial improvements in glycemic management with A1c reductions of 1.5% and Time in Range increases of 3.3 hours/day to 53% at three months and that this improvement was maintained out to one year. Turning to other pediatric AID users, results from the GIF study (NCT04269668) demonstrated that MiniMed 780G drove significantly greater Time in Range improvements and helped 71% of participants meet consensus Time in Range targets. Seeing the strong results from adult AID trials maintained among pediatric patients has us reflecting on the efficacy of AID algorithms and their ability to adapt to patient needs and behaviors to help drive improvements in glycemic management for such varied patient populations.

• AID use among very young children has shown incredibly positive results. Of note, data out of the University of Cambridge on the KidsAPS02 study, which is evaluating the use of the CamAPS FX AID system among pediatric patients ages 1-7, demonstrated 72% Time in Range among AID users compared to 63% for those on sensor-augmented pump therapy. Additionally, post-hoc results of Insulet’s Omnipod 5 preschool trial were presented and demonstrated the value of adjustable insulin targets among young children to take into account activity and changing eating behaviors. Across these studies, we see a growing body of evidence in favor of AID use, even among very young children, and hope that AID can become more broadly available for use in this age group soon. Based on the data we've seen, we imagine it will have significant potential to reduce parent and caregiver worry and improve diabetes management for young children.

• Data from adult patients transitioning directly from MDI to MiniMed 780G demonstrated remarkable improvements in Time in Range highlighting another population where AID is continuing to drive improved glycemic management. Specifically, adults who transitioned from MRI saw a 1.4% A1c reduction at six months that was coupled with a Time in Range increase of 6.6 hours/day to 71%. As conversations around who is a good candidate for diabetes technology continue, we are encouraged to see such strong improvements among prior MDI users as we know that a large percentage of AID users updated to these systems as prior pump users. However, this new data demonstrates that direct transitions from MDI may be able to help accelerate improvements in glycemic management, and we also see this data as a good reminder that technologies should be offered to a wide variety of patients allowing patients and providers to collaboratively choose the best diabetes management tools for each individual. Of course, in the US, the “choices” are also sometimes made by payers so that people with diabetes may have more limited choices – still, we’d say that having more choices even just for payers is very helpful.

7. Put it in the blender: Tech and therapy no longer existing as distinct entities; greater discussion of blending technology and therapy to improve patient outcomes

The blending of tech and therapy was omnipresent at this year’s conference, from greater use of CGM in clinical trials to symposia discussing both new technology and therapies. For example, the Lilly symposium highlighted the Tempo Smart Button and the new rapid acting insulin Lyumjev on equal footing and spent significant time discussing how the two might interact. Using the Tempo Smart Button, patients can automatically record their Lyumjev data (dose quantity, day, time) in a partner app connected via Bluetooth. The ability to use this rapid acting insulin, which is indicated to improve glycemic control through subcutaneous injection either at the start of the meal or within 20 minutes of eating (!), in combination with cutting edge technology is contributing to the “insulin evolution.” Novo Nordisk and Sanofi also had a major focus on their smart pen products to promote improved patient experience in their exhibit hall booths. Taking a step back, we were thrilled to see more discussion of therapy at this conference than has historically been the case and we are pleased to see that people are thinking about technology and therapeutics more jointly, as both entities are essential to achieving positive patient outcomes.

• We also were so pleased to see data utilizing CGM in clinical trials across the board. The SURPASS-3 CGM trial utilized CGM to assess the glycemic variability of tirzepatide compared to insulin degludec, and found that the dual GIP/GLP-1 agonist led to an impressive 6 hour/day improvement in Time in Tight Range compared to insulin. Moreover, although we have yet to see the data, the OWNARDS 2 study with insulin icodec also utilized CGM to assess the once-weekly basal insulin’s effect on glycemic variability and hypoglycemia. Of course, we also spent substantial time thinking about the future role of CGM in clinical trials following the International Consensus Meeting, and we are so excited to see this overlap between technology and therapy continue to grow.
• We were enthralled by a trial evaluating the combination of SGLT-2s and AID technology in type 1 diabetes. Coming out of the prestigious UVA Center of Diabetes Technology, the CiQ-SGLT2 study was read out by UVA’s own Dr. Jose Garcia-Tirado to a packed ATTD 2022 audience. Overall, the eight-week RCT found that combining low-dose SGLT-2 (empagliflozin) with Control-IQ improves Time in Range +2.4 hours/day to 81% for adult type 1s (n=32) vs. AID alone. Notably, this Time in Range benefit was gained without an increase in Time Below Range. However, two participants on empagliflozin discontinued treatment due to diabetic ketoacidosis (DKA) and 13 out of 18 participants on empagliflozin experienced ketosis without DKA, compared to three out of 17 participants on only Control-IQ. While the prevalence of ketosis without DKA among empagliflozin users is concerning, we remain optimistic that the advancing field of continuous ketone monitoring may make SGLT-2 therapy for type 1s safer and more accessible.

8. The next frontier in glycemic control and obesity: Tirzepatide touted as the true “game changer” in T2D management with jaw-dropping weight-loss results in SURMOUNT-1

Dual GIP/GLP-1 receptor agonist tirzepatide stole the show in therapy during ATTD, between the riveting KOL comments in a session on dual agonists and the remarkable results of the SURMOUNT-1 trial announced in the wee hours of the night after Day #1. We were moved by an all-star KOL panel that highlighted the SURPASS-3 CGM results, discussed the patient-centered experience on tirzepatide, and even speculated on the potential to achieve diabetes remission with tirzepatide. We could not be more excited about the potential for this dual agonist to transform treatment of type 2 diabetes through its strong glycemic control, with the added benefit of 20%-plus weight loss. Dr. Julio Rosenstock (University of Texas Southwestern) summed up the dual GIP/GLP-1 agent’s clinical profile, stating that “tirzepatide has moved the goalpost for type 2 diabetes management towards attaining ‘diabetes reversal or remission,’ which may no longer be the ‘impossible dream to reach the unreachable star’!” Though the SURMOUNT-1 results were not officially discussed at ATTD, we’d be remiss not to mention the highly impressive findings from this hallmark study in people with obesity. Participants on tirzepatide 15 mg achieved a whopping 22.5% weight loss on average over the 72-week trial, with 63% of participants on tirzepatide 15 mg losing at least 20% of their body weight compared to just 1% of those in the placebo group. We’re especially keen to see the full results in those with prediabetes, who will continue treatment for an additional two years. As we noted in our 2021 Reflections, tirzepatide will play a key role in filling the obesity treatment gap and bringing weight loss within reach for the millions of people living with overweight and obesity. Per Lilly’s 1Q22 update (also during ATTD week), the company is re-opening discussions with the FDA to explore options for regulatory submission that include an expedited approval process. We’ll be following updates, and we look forward to seeing additional SURMOUNT-1 data during the many upcoming conferences this summer.

Glucose Monitoring Highlights

Dexcom G7’s European launch is underway in the UK: Dr. Lala Leelarathna and Dr. Hood Thabit on their experiences supporting the first seven people in the world using Dexcom G7 and these early users’ feedback

In major news, Dexcom G7 has officially launched in the UK, where University of Manchester’s Dr. Lala Leelarathna and Dr. Hood Thabit have become some of the world’s first top clinician’s first seven people in the world using G7 over the last few weeks. This follows Dexcom G7’s CE-Marking in March and reflects Dr. Thabit’s commentary at Diabetes UK 2022 in late March, where he shared that his patients in the UK would be among the first in the world to use G7. During today’s action-packed Dexcom-sponsored symposium, Dr. Leelarathna and Dr. Thabit discussed their early experiences with the system and highlighted their patients’ and their favorite features of the system. Overall, Dr. Leelarathna characterized the early users’ feedback as “really positive” and “really encouraging” both in terms of design usability and performance and that these seven folks are “very thankful” to be the first to use this next-gen system.

Digging into the details, Dr. Thabit highlighted several features with which the early users are particularly thrilled, some of which were new to us (see below!) and we’re also now hugely excited:

• The minimal size that means they “hardly notice they’re wearing it”;
• the easier insertion process (we note it’s also less environmentally damaging);
• the smoother onboarding process (it may be that clinicians are even more particularly thrilled than patients, who may not have anything to compare);
• the faster warmup period (30 minutes vs 120 minutes);
• he integration of the CLARITY app into the G7 app (we have been waiting for this forever and hadn’t remembered this!);
• the sensor’s accuracy (MARD of 8.2% in the pivotal trial) – while we don’t think the accuracy of Dexcom has worried many individuals since the G4 was approved- even while the Dexcom’s first-gen SLS had accuracy issues, continuous trumps fingersticks any day in terms of better and easier, if not yet cheaper; and
• the more flexible alarm settings, including delayed high alarms to reduce burden and the ability to turn off all alarms for six hours,fter several confirmations that that decision is intentional. While we know that multiple patients like this decision, we’re personally slightly surprised that Dexcom opted for this, given that over many years, they have stuck to a “true North” of keeping the <54 alarm on, just in case. Ah well, it’s good to see adaptation – we can only imagine how well-researched this question was!

Separately, Dr. Nicholas Argento (Maryland Endocrine and Diabetes Center) discussed his favorite features of the Dexcom G7 after reviewing the outcomes of the pivotal, which was published in DT&T in February. These included several of which we weren’t previously aware and that were not explicitly mentioned by Dr. Leelarantha and Dr. Thabit, including:

• An overpatch that ensured virtually full sensor use during the clinical trial (practically no sensor loss);and
• that sensor insertion automatically starts the warmup process (people on our team who wear Dexcom have often forgotten to press go, and then the warmup process has been delayed, sometimes for hours); and
• Some also like that the warmup can definitely be overlapped with the previous sensor worn in a 12-hour grace period beyond 10 days (as we previously had presumed but hadn’t confirmed) though that one makes us more worried since we know multiple patients that have had problems with this in the past given that the sensors do run out of juice, figuratively and literally – they are not made to be extended We’d love to see a robust patient access program started and expanded, given that the margins have expanded on these sensors, which is perhaps the best factor of all – there’s more room to drop the price, particularly imagining how much volume will be expanding.

It is hugely exciting to hear about these early experiences – seven – and we look so forward to seeing additional real-world data on productivity (steming from quality of life) and user satisfaction soon!

• As a side note during his presentation, Dr. Argento shared hope that Dexcom G7 will be cleared by the FDA “before ADA” but was hesitant, stating “we’ll see.” As a reminder, Dexcom submitted G7 to the FDA in 4Q21, as CEO Kevin Sayer shared during his JPM 2022 presentation. This submission is based on the pivotal data that was published in DT&T in February. Given these early patient experiences and the impressive pivotal data, we cannot wait for G7 to become available in the US and certainly would be thrilled if it were cleared ahead of ADA, which is just over a month away – that said, as noted earlier, that’s probably not going to happen though yet and still, we certainly have the sense, stemming from Dexcom’s 1Q22 call on Thursday, that this was possible but unlikely. Indeed, given the well-reported delays at the FDA, we wouldn’t shocked if it took longer, largely or solely due to FDA capacity. We also will be eager to hear more on where the interoperability conversations are going – to soon to say much on this front, but it’s certainly not uncomplicated.
• In Dr. Argento’s slide summarizing the clinical data on Dexcom G7, he included data on the “original CGM algorithm,” which was used to support the CE-Marking and showed a MARD of 9.5%. This is the first we’ve seen this data, and while still a solid MARD, we’d note that this is slightly higher than that achieved in the pivotal. The study included 98 participants (ages 2+) with T1D or T2D. This compares to 316 participants in the adult pivotal trial and 132 in the pediatric pivotal trial, which could possibly explain the difference in performance, though we are not exactly sure about that. Although he included the data, Dr. Argento was clear that he wasn’t intending to discuss that previous study during his presentation nor in the Q&A.

“It is time for a systematic approach to using CGM to adjust type 2 diabetes management in insulin using patients”: Dr. Rich Bergenstal presents novel C²GM Therapy Adjustment Guide

Dr. Rich Bergenstal (International Diabetes Center) took to the stage to deliver one of the most engaging presentations of ATTD 2022 calling for greater action based on CGM data and presenting his novel C²GM (CGM Clinician Guided Management) treatment algorithm. Dr. Bergenstal began his presentation recognizing the last ten years in the adoption of Time in Range – we commend Dr. Bergenstal on the impact he and IDC have had throughout this time. In this talk, he also asserted that some of the field is currently stuck at an “analyze” stage with many clinicians, especially those in primary care, struggling to act on CGM data to optimize glucose management.  Population level glycemic management as well as data from trials like MOBILE demonstrating a lack of CGM-driven clinical action certainly support this sentiment, although we’d say MOBILE was designed to show changes from behavior specifically and not really the impact from therapeutic changes – i.e., in some trials, the design effectively understated the changes that could be made. One could also argue, of course, that RCTs provide more motivation to many PWD through the more thorough care from clinicians helping in the trial.

Additionally, Dr. Bergenstal lamented current best practices in the treatment of type 2 diabetes such as “treat to target” and “fix fasting first,” both of which he expressed actually involve substantial time on the part of providers to walk through current treatment algorithms and calculations including checking for over-basalization. Moving on from current treatment, Dr. Bergenstal presented a novel paradigm for CGM-based management of type 2 diabetes focusing on Time in Range and Time Below Range. In this paradigm, providers walk through a three step process to: (i) determine if a patient has comorbidities where a GLP-1 or SGLT-2 should be considered; (ii) find the % Time in Range and % Time Below Range on the patient’s AGP and ask “is the Time in Range >70%” and “is the Time below Range <2%”?; and (iii) find the Time in Range/Time Below Range category in the treatment algorithm table that corresponds to the patient’s CGM data and adjust treatment as necessary. We love this – so action-oriented – and were absolutely thrilled to see it, since we also think it’s so easy that it can also be explained to patients and recommended that they tell their providers about it if they do not think their providers are working in this way. As providers are only using two metrics from a patient AGP, this creates four simple categories that patients will fall into and for which Dr. Bergenstal’s group has developed simple and specific treatment adjustments allowing providers (as well as patients who have health literacy) to take action quickly and easily to help improve glycemic management. The four categories and immediate goals for therapy adjustment are as follows:

• Time in Range >70% and Time Below Range <2%: Continue regiment by continuing to optimize current therapy and reinforce lifestyle changes and taking insulin as prescribed. Recommended follow-up in 3-4 months.
• Time in Range >70% and Time Below Range >2%: Address hypoglycemia and stop sulphonylurea use if present and reduce background insulin by 10% if Time Below Range is 8-12% or by 15% if Time Below Range is >12%. If the patient is not on a sulphonylurea, decrease the total background insulin dose by 10% if Time Below Range is 2-7%, by 15% if Time Below Range is 8-12%, and by 20% if Time Below Range is >12%. Recommended follow-up of 2 weeks.
• Time in Range ≤70% and Time Below Range <2%: Address hyperglycemia and consider adding or adjusting GLP-1 therapy, otherwise increase background insulin dose by 10% if Time in Range is 51-70%, by 15% if Time in Range is 30-50%, and by 20% if Time in Range is <30%. However, if the patient experiences nocturnal hypoglycemia, consider a smaller increase in insulin dose. Recommended follow-up of 2 weeks.
• Time in Range ≤70% and Time Below Range >2%: Address hypoglycemia immediately and consider referral to a diabetes educator or specialist and stop sulphonylurea use if present and reduce background insulin by 10% if Time Below Range is 8-12% or by 15% if Time Below Range is >12%. If the patient is not on a sulphonylurea, decrease the total background insulin dose by 10% if Time Below Range is 2-7%, by 15% if Time Below Range is 8-12%, and by 20% if Time Below Range is >12%. Additionally, the patient should be referred to a diabetes educator for options to treat hyperglycemia including the potential addition of a GLP-1 or mealtime insulin. Recommended follow-up of 2 weeks.

We see the simplicity of this system as a major, major win for providers with busy schedules (i.e., so many!) or who may be currently less familiar with diabetes treatment, allowing them greater time to learn from patients and engage in collaborative decision-making. Additionally, we imagine the breakdown of recommendations by patient Time in Range will be incredibly beneficial for providers by doing much of the necessary data interpretation for them.

• Dr. Bergenstal’s group came to the decision to use ≤2% Time Below Range as the threshold for clinical adjustment category based on analysis of participant AGPs from the MOBILE study. While this is certainly lower than the Time in Range consensus target for <4% Time Below Range, Dr. Bergenstal explained that he and his collaborators evaluated every available AGP from MOBILE study participants and arranged them based on Time Below Range which varied from 0% to 9%. Of these patients, Dr. Bergenstal and his group came to consensus on whether or not they thought it would be safe to intensify treatment for each percentage increase in Time Below Range ultimately deciding that they were only comfortable intensifying treatment for those with Time Below Range <2%.
• Dr. Thomas Martens (International Diabetes Center) presented case study applications for the C²GM treatment algorithm using participants from the MOBILE study. Dr. Martens created four composite AGP profiles of MOBILE participants such that one composite AGP profile fell into each treatment adjustment category. The first composite profile (n=20) demonstrated a Time in Range of 80% and Time Below Range of 0% leading Dr. Martens to explain that adjustments to therapies were not needed and that providers could comfortably and safely follow-up with these patients 3-4 months later. The second composite profile (n=5) had a Time in Range of 76% and a Time Below Range of 5% for which Dr. Martens recommended immediately addressing hypoglycemia and discontinuing the use of sulphonylureas if present. Importantly, Dr. Martens encouraged providers to follow-up with these patients two weeks later in case these changes impact Time in Range, which could then be addressed in a subsequent appointment. The third composite profile (n=139) demonstrated a Time in Range of 34% and a Time Below Range of 0% and represented the majority of the patients included in this analysis. For this profile, Dr. Martens identified addressing hyperglycemia as the most important treatment goal and advised adding or adjusting GLP-1 therapy or increasing background insulin. Finally, the fourth composite profile (n=12) demonstrated a Time in Range of 51% and a Time Below Range of 5%. Since these patients struggled with both hyper and hypoglycemia, Dr. Martens advocated for a coordinated care-team approach with referral from primary care to a diabetes educator or other specialist. However, in the immediate, Dr. Martens also stressed the importance of addressing hypoglycemia and taking these patients off any sulphonylureas and reducing background insulin to prevent any potentially life-threatening hypoglycemic episodes.

FreeStyle Libre 3 US accuracy trial (n=100 adults and pediatrics): MARD of 7.8% and ±20/20% of 93% vs. YSI (n=6,845 paired points); submitted to FDA in 2021

In an exciting update during an Abbott symposium, Dr. Emma Wilmot (University Hospitals of Derby and Burton) read out data from an accuracy trial for FreeStyle Libre 3 (n=100), reporting a topline MARD of 7.8%. Before reading out data from the trial, Dr. Wilmot confirmed that this study, conducted across four centers in the US, was used to support Abbott’s FDA submission of FreeStyle Libre 3 as an iCGM, which happened sometime in 2021. However, we do not believe that the data presented came from the US pivotal trial.

This accuracy study enrolled participants over four years old with type 1 (T1D) or type 2 diabetes (T2D) on insulin therapy. The study included five people younger than six years old, 39 people between six and 17 years old, and 56 people ≥18 years old. The 95 participants who were ≥six years old had venous blood glucose analyzed over three separate in-clinic visits using the Yellow Springs Instrument Life Sciences 2300 STAT Plus YSI analyzer. For the five participants under six years old, BGM was used as the study comparator per the FDA’s preference, as we understand it. Notably, this study did not include deliberate glucose manipulation, a technique that is typically used in a pivotal study to ensure enough paired points are generated in the hypoglycemic range.

We first learned of the system’s 7.8% MARD yesterday when Abbott announced a new partnership with Ypsomed and CamDiab to incorporate FreeStyle Libre 3 with the myLife CamAPS FX AID system by the “end of 2022,” which could make myLife CamAPS FX the first commercial system to come to market with an Abbott CGM.

• In the study, 93% of matched CGM-YSI pairs met the ±20/20% overall agreement rate. This agreement rate, which is generally a more robust metric than overall MARD, places the FreeStyle Libre 3 sensor accuracy right around that of Dexcom G7. Again, for reference, Dexcom G7 reported a ±20/20% agreement rate of 95% and an overall MARD of 8.2% for upper arm wear and a ±20/20% agreement rate of 93% and an overall MARD of 9.1% for abdomen wear.

Glucose Concentration	Within ±15 mg/dL	Within ±20 mg/dL	Within ±40 mg/dL
<70 mg/dL	85.8%	93.3%	99.2%
	Within ±15%	Within ±20%	Within ±40%
≥70 mg/dL	87.8%	93.4%	99.5%
	Within ±20 mg/dL and within 20% of reference
All results	93.4%

• By day, FreeStyle Libre 3 demonstrated the strongest accuracy on days nine through 12 with the weakest accuracy on days seven and eight of sensor wear. While this sounds slightly surprising, since historically, the first couple of days are the least accurate – the MARD on the earliest days were virtually the same as Days 7 – 8. Specifically, MARD on the first day of sensor wear was 8.6%, staying relatively flat through days seven and eight at 8.7%. On days nine through 12, the MARD dropped significantly, to 6.4%, increasing to 7.0% through days 13 and 14. The trend of increasing accuracy after day one is common among CGMs and seen in all systems currently on the market.

Wear period	Within ±20 mg/dL and within 20% of reference	MARD
Beginning (Days 1 - 3)	92.1%	8.6%
Early middle (Days 7 - 8)	91.3%	8.7%
Late middle (Days 9 - 12)	96.0%	6.4%
End (Days 13 - 14)	95.0%	7.0%
Overall	93.4%	7.8%

• Across glucose ranges, quite interestingly, FreeStyle Libre 3 demonstrated strong accuracy and generated the most accurate readings in the high ranges of >350 mg/dL and 251-350 mg/dL. Specifically, sensor MAD (mean absolute difference) was 16.5 mg/dL in the range of <54 mg/dL and 8.0 mg/dL in the range of 54-69 mg/dL. When in range (70-180 mg/dL), MARD was 8.4%. Turning to non-severe hyperglycemia, MARD was impressively low at 6.3% in the range of 181-250 mg/dL. While we find this data encouraging overall as accurate CGM data in both hypo and hyperglycemia is especially important to inform treatment decisions, hyper-accuracy is not quite as important as it used to be in the “before-arrow” era. While it remains to be seen how clinically relevant FreeStyle Libre 3’s lower MARD will be, compared to other features such as real-time alarms, data streaming, and the sensor’s small size. The degree to which various factors represent key differentiators varies – some  of the other factors may mean significantly more to patients than a an improvement in sensor accuracy that they may not perceive as significant.

Glucose Level	MAD	MARD
<54 mg/dL	16.5 mg/dL
54-69 mg/dL	8.0 mg/dL
70-180 mg/dL (in Range)		8.4%
181-250 mg/dL		6.3%
251-350 mg/dL		4.9%
>350 mg/dL		4.1%

Dexcom G7 safety and accuracy study in pediatric type 1s: MARD of 8.3% and 9.0% for children aged 7-17 (n=132) on arm and abdomen, respectively; MARD of 9.3% for children aged 2-6 (n=32) overall; study likely to support G7 pediatric type 1 FDA submission

During a highly attended Dexcom symposium, Joslin’s Dr. Lori Laffel presented  results from a prospective, multicenter, single-arm Dexcom G7 safety and accuracy study in children aged two through 17 (n=164 children with 240 sensors). This presentation follows the CE-Marking of Dexcom G7 in March 2022, as well as the publication of the adult pivotal trial results in DT&T in February 2022. Dexcom G7 was submitted to the FDA in 4Q21 as an iCGM. While management has yet to provide specific timelines for expected clearance, at JPM 2022 as well as during 1Q22 reporting, Dexcom CEO Mr. Kevin Sayer conveyed a hope and expectation for  clearance and launch in the US in 2022. During 1Q22 results reporting, he conveyed that communication was going well and that although approval before ADA 2022 in early June was not likely, the regulatory work was going well. That approval before ADA isn’t expected wasn’t too surprising from our view, since although we don’t know the exact date the submission was, ADA is extremely early this year. Although Dr. Laffel did not directly refer to this study as a pediatric pivotal trial for Dexcom G7, we assume (and recognize this is conjecture) that this study could be used to support an FDA indication for pediatric type 1s, given its similar design to the adult pivotal trial.

The study was conducted in children (n=132, ages 7-17) and very young children (n=32, ages 2-6) with type 1 diabetes who wore G7 sensors on the upper arm and abdomen, as well as the upper buttocks in the case of the very young children. Data were collected from a total of 240 sensors for 10.5 days each to evaluate the 10-day wear of the G7 sensor along with the 12-hour grace period. Participants conducted either two clinic visits (13-17 years old) or one clinic visit (7-12 years old) on day one or two, four or seven, and 10 or 10.5 to collect YSI blood glucose measurements. During the in-clinic sessions, participants’ glucose levels were manipulated from over a specified range and YSI values were measured every 10-15 ± 5 minutes. Very young participants (2-6 years old) had only one clinic session for four hours, during which BGM readings were obtained with the Ascensia CONTOUR NEXT BGM as a comparator, as YSI wasn’t an option in this age group. While Dr. Laffel did not provide any baseline characteristics on gender or race/ethnicity (she had to get through all the data quickly!), we will write to ask for details on this front. While significant diversity in race/ethnicity wouldn’t be expected since T1D is not experienced disproportionately more in people of color, we are always interested to see diversity in other respects that are always conveyed – in a larger trial, sub-group parameters like household income, insurance, etc. would be terrific to see in terms of higher or lower TIR, though we imagine a trial with 100 PWD wouldn’t be powered to show significance.

• Interestingly, Dexcom G7 had a MARD of 8.3% for arm wear and 9.0% for abdomen wear in children aged 7-17, and an overall MARD of 9.3% for children aged 2-6. MARD values were calculated by comparing CGM and in-clinic YSI glucose measurements for a total of 15,437 matched CGM and YSI data pairs. These pairs were then analyzed to establish the percentage of pairs within ±15/15% and ±20/20%. As background, these measurements reflect the proportion of CGM values within 15 mg/dL and 20 mg/dL of YSI values <80 mg/dL and within 15% and 20% of YSI values >80 mg/dL. For upper arm wear in children aged 7-17, the overall ±20/20% agreement rate was 95%, and for abdomen wear, the overall ±20/20% agreement rate was 93%. Overall, in children aged 2-6, the ±20/20% agreement rate was 92%. It’s worth noting that these values are quite strong and well within the FDA special controls necessary for clearance as an iCGM device. Additionally, the strong MARD data gives G7 a strong advantage among other FDA cleared CGMs for children and very young children.

Placement	Overall ±20/20% Agreement Rate	MARD
Upper Arm, 7–17-year-olds	95%	8.3%
Abdomen, 7–17-year-olds	93%	9.0%
Overall, 2–6-year-olds	92%	9.3%

• Across glucose ranges, G7 demonstrated strong accuracy in 7–17-year-olds, (n=122) with strongest performance between 61 - 80 mg/dL and 301 - 400 mg/dL. Specifically, sensor MAD (mean absolute difference; measurement used instead of MARD for glucose <80 mg/dL) was 11.3 mg/dL in the range of 40 - 60 mg/dL and 6.4 mg/dL in the range of 61 - 80 mg/dL. Between 81 and 180 mg/dL, MARD was 8.4%. Turning to hyperglycemia, MARD was impressively low at 7.6% in the range of 181 - 300 mg/dL and 5.4% in the range of 301 - 400 mg/dL.

• By day, G7 demonstrated the strongest accuracy in 7 - 17-year-olds (n=122) on days four through ten with the weakest accuracy on day one of sensor wear. Specifically, MARD on the first day of sensor wear was 11.7%, but quickly fell and was calculated at 7.9% by day two. Additionally, accuracy was maintained throughout the life of the sensor with a MARD of 7.3% at day 10.5 (10.5!). The trend of lower accuracy on day one is common among CGMs and seen in virtually all systems currently on the market, though there has been much improvement over time in all the systems as well.

• Notably, Dr. Laffel stressed that in her experience, beyond the trial results, Dexcom G7 marks a particularly notable improvement for her pediatric patient compared to Dexcom G6. Specifically, Dr. Laffel emphasized that G7 is significantly smaller than G6, so she believes many will be  more amenable to the limited skin “real-estate” that many of her younger patients often have – or, want to give up for a medical device. As well, Dr. Laffel noted that the integrated transmitter and sensor will be a game changer, considering that many of her younger patients would accidentally throw out the transmitter by accident at the end of sensor wear.
• Strong patient sentiments shown on the G7: Although we haven’t yet been able to get sample sizes, Dr. Laffel also presented data from a study showing that 96% of surveyed pediatric participants reported the Dexcom G7 insertion process as “easy” or “very easy.” To boot,  75% reported that the insertion process and comfort level of G7 had improved compared to G6. The G6 insertion, of course, was itself a major improvement over G5 - see more here from diaTribe for a quick snapshot of the early progress shown for the G5 back in 2016. While the authors didn’t note anything special about more challenging insertion for G5 at that time, when the G6 was approved in 2018, the headline of diaTribe’s article was titled “Dexcom G6 Review: No Fingersticks CGM, One-Button Insertion, and 10-Day Wear” (our emphasis).

Retrospective, observational chart review in broad type 2 population (n=2,331) at International Diabetes Center: CGM use correlated with 0.9% A1c reduction from 8.9% to 8.0%; fivefold (!) increase in participants taking zero medications, from 5% to 25% of cohort

During an oral presentation session, Dr. Anders Carlson shared data from the International Diabetes Center highlighting CGM use in a broad population of people with type 2 diabetes. To start, Dr. Carlson noted that most studies evaluating CGM use in type 2s examine people on insulin therapy (e.g., the MOBILE RCT), and that large prospective randomized studies of CGM use in non-insulin-treated type 2s are “sparse.” Dr. Carlson did point out that in one such study, albeit a small one (Wada et al. 2020), CGM use was associated with a 0.5% A1c decrease in non-insulin-treated type 2s (n=49) compared to those on BGM (n=51). Turning to the IDC study, Dr. Carlson explained that he and his colleagues performed a retrospective, observational EHR and claims review studying all people with type 2 diabetes who received diabetes care and had insurance through IDC’s parent integrated health system, HealthPartners (n=23,843) between January 2018 and December 2021. Within this population of type 2s, roughly 10% (n=2,331) received a new CGM order, and a majority of participants filled their prescriptions within 30 days of receipt (84%) and ultimately got a FreeStyle Libre (89%) or Dexcom (11%) CGM. Overall, 93% of CGM orders were filled.

• Across the entire type 2 cohort, using CGM use was correlated with a 0.9% decrease in A1c, from 8.9% at baseline, on average, to 8.0% post-CGM (p<0.0001). Dr. Carlson explained that the “baseline CGM value” was the closest A1c between zero and six months prior to starting CGM, and that the “post-CGM A1c value” was the closest A1c between eight weeks and 12 months after using CGM. Notably, the percentage of participants with an A1c under 8% increased by 17%, from 36% at baseline to 53% after CGM use. Dr. Carlson noted that this metric is worth looking at since in many places, including his home state of Minnesota, diabetes care quality metrics include if HCPs meet a treatment target of A1c <8% and it is currently the HEDIS criteria for diabetes management. He also noted, of course, how much better this was for PWD! Dr. Carlson then turned to a graph (see below) showing the A1c change among participants with a baseline A1c between 8% and 10%, with bubble sizes proportional to the number of patients. As Dr. Carlson noted, it’s interesting to see that while CGM did not bring down A1c for every single participant, it did for most.

• After initiating CGM, Dr. Carlson stressed that the number of individuals in the study taking no medications increased fivefold, from 5% to 25%. According to Dr. Carlson, this result indicates that the mean 0.9% A1c improvement after using CGM likely stemmed from improved nutrition and lifestyle changes, as opposed to pharmacotherapy intensification. The percentage of participants who were taking ≤two medications increased as well, from 62% at baseline to 71% after CGM use. While from our view, less medicine isn’t better for everyone, presumably IDC has ways to determine who should be on a therapy like GLP-1 or SGLT-2s to reduce CV or kidney disease risk, rather than for glycemic management. And indeed, separately from Dr. Carlson, we learned that the observed decrease in pharmacotherapy stemmed almost all from a decrease in analog insulin, sulfonylurea, and DPP-4i usage. For example, SFU use went down from 27% to 16%.
  
  • Insulin usage among participants decreased from 61% pre-CGM to 50% post-CGM, which we find most interesting as it seems to suggest that CGM can drive meaningful lifestyle changes that reduce insulin dosing as well as dependence. Ultimately, we’d love to see Time in Range associated with this population as well as more granular CGM tracings to understand the needs better.
  • Interestingly, GLP-1 usage in the cohort increased regardless of baseline A1c. For those with an A1c over 10%, analog insulin use increased, whereas for those with a baseline A1c ≤10%, sulfonylurea use decreased. At the conclusion of his presentation, Dr. Carlson argued that these data suggest wider CGM coverage in a type 2 population can be beneficial, and he called for more research into CGM use in non-insulin-treated type 2s. We certainly understand the need and call for this.
• Dr. Carlson stressed that at the beginning of the study in 2018, CGM was available to all type 2s insured by the IDC, which may account for the larger number of people not on insulin and on fewer medications in the study. We wonder whether CGM is still offered to all type 2s who are insured through IDC, and if not, we would be eager to learn why this is not the case.
• There was no correlation between ethnicity/race and A1c. This is especially worth noting because the cohort was relatively diverse (1% Native American/Alaska Native, 8% Asian, 18% Black, <1% Native Hawaiian or Other Pacific Islander, 59% White, and 14% Unknown/Other). There were also no correlations between A1c and BMI or the number of concurrent medications. However, male sex (p=0.03), younger age (p=0.001), and filling CGM within 30 days (p=0.0003) were all associated with a lower A1c.

Dr. Tadej Battelino unveils plans for a new consensus statement on the use of CGM metrics for therapy intensification in type 2 diabetes – including those not on insulin – with aim for completion by February 2023

Closing out this morning’s ATTD-Abbott School, Dr. Tadej Battelino (University Medical Center Ljubljana, Slovenia) discussed yesterday’s consensus meetings on: (i) the use of CGM metrics in non-device clinical trials; and (ii) the use of CGM metrics for therapy intensification in type 2 diabetes. See here for our report on the former meeting, which included representatives from industry, ADA, JDRF, NIH, FDA, CDISC, the Helmsley Charitable Trust, and ATTD, as well as many clinicians and researchers. While he didn’t delve into details on yesterday’s meeting, Dr. Battelino did share that the meeting on CGM in clinical trials effectively updated a draft of the consensus report and importantly, added recommendations on how to statistically present CGM data and what a noninferiority margin should be for various CGM metrics in clinical trials.

While we attended the former meeting on CGM metrics in non-device clinical trials, the latter, which was also held yesterday, was only open to clinicians and researchers. This initial meeting aimed to enable discussion of and agreement upon the report’s goals, sections, and vision, which will guide the drafting of the consensus recommendations as the next step. Based on Dr. Battelino’s commentary, the report will offer guidance on the use of CGM data to guide treatment decisions for those on insulin as well as those who are not on insulin, which Dr. Battelino characterized as “the tougher part.” Dr. Battelino noted his desire for the group to come to consensus that at least intermittent CGM should be used for all people with type 2 diabetes to drive treatment decisions; however, he did not disclose how likely the group is to make such a recommendation. He also hopes that the group comes to consensus on the threshold at which someone should be using personal CGM continuously and the CGM metric thresholds at which someone should initiate insulin, which is incredibly important to combating clinical inertia and providers’ hesitancy to initiate insulin. The group aims to complete the consensus report by February 2023, likely in time for ATTD 2023, which will be held on February 22-25, 2023 in Berlin.

• Thus far, the type 2 consensus report process has been endorsed by the EASD, the ADA endorsement process has been initiated, and a Japanese diabetes group has been invited to participate, ensuring that the consensus will be more global than simply the US and/or Europe. On industry participation, Dr. Battelino drew a comparison between the two meetings, noting that unlike the clinical trials consensus report, the type 2 therapy intensification consensus report will have no industry sponsorship or involvement, as it is intended for use by clinicians rather than in clinical trials, many of which involve industry members.
• This type 2 treatment intensification consensus report is desperately needed to combat clinical inertia, to support clinicians in making better informed type 2 treatment decisions, and to create treatment guidelines specific to the growing and heterogenous population of people with type 2 diabetes on CGM. The support for CGM in type 2 diabetes continues to grow. Last year’s ATTD sessions saw the readout and publication of MOBILE study, which drove the ADA to recommend CGM for anyone on insulin. Not only has support for CGM in those using insulin grown, but just last week, we learned that a whopping 40% of Abbott’s 4 million FreeStyle Libre users (i.e., 1.6 million people) are type 2s not on basal-bolus therapy. Given the increasing use of CGM in type 2 diabetes, there a tremendous need to create guidelines that inform clinical decisions driven by these tremendous devices. The need for guidelines is particularly high because so many type 2s receive their diabetes care from primary care providers, who may be less familiar with CGM technology, CGM metrics, and CGM-guided treatment decision-making. Given this massive gap in current guidelines, we applaud the consensus group for its efforts and look forward to seeing its recommendations, particularly those on the CGM metric thresholds or patterns that suggest treatment should be changed and those that offer guidelines for supporting those not on insulin based on CGM data.

Retrospective payer claims analysis (n=700+) shows professional CGM use associated with a -0.5% A1c improvement in type 2s on multiple non-insulin diabetes medications; professional CGM use associated with increased insulin, GLP-1, and SGLT-2 initiation, but rates still low

In an afternoon oral presentation session, Ms. Poorva Nemlekar (Lead Health Economics and Outcomes Research Specialist in Global Access, Dexcom) presented a large and impressive retrospective RWE analysis. We were thrilled to see more focus on professional CGM initiation, as we believe that it is an often-under-utilized resource that can drive cost-savings for so many individuals. In the analysis, professional CGM was used in a very specific population: people with type 2 diabetes on more than two non-insulin diabetes medications who are not achieving A1c targets. Professional CGM was associated with a -0.5% A1c improvement in this population, compared to patients who did not use professional CGM, which we see as quite meaningful – we have learned extensively from dQ&A about various people with type 2 diabetes who are not at their target A1cs, even people on medications like GLP-1s that are often associated with very positive outcomes in RCTs. And of course, this shouldn’t be a surprise – they are just in a different stage of their disease duration and likely have less beta cell function. But, it is surprising the number of those PWD who have either not had the advice to intensify their medicine or just have not done it.

The analysis included n=707 adults ages ≥30 (average 66 years) with baseline A1c values between 7.8% and 10.5% (average 8.7%) with at least one claim of professional CGM use between January 2018 and October 2020 but no prior professional or personal CGM use. Nearly three-fourths of the group was on Medicare (72%), and 50% were non-Hispanic White, making this analysis a more diverse one in terms of participant payor and ethnicity than is often seen in diabetes technology studies where the population is often made up of >90% those who are non-Hispanic White and who may have commercial insurance. These 707 participants were compared to a cohort of 14,774 type 2s on ≥2 non-insulin medications who had never used CGM and did not initiate professional or personal CGM between January 2018 and October 2020. To assess the impact of professional CGM, the researchers compared A1c and medication data six months prior and six months after professional CGM initiation (in the professional CGM group) or oral diabetes medication initiation (in the non-CGM group).

• Those who initiated professional CGM saw a -0.5% A1c improvement compared to their non-CGM using counterparts when adjusted for baseline A1c (p<0.0001). Specifically, those who initiated professional CGM saw a -0.8% A1c improvement from 8.7% six months prior to CGM initiation to 7.9% six months after initiation, whereas those who did not use professional CGM saw a 0.3% A1c reduction from 8.5% to 8.2%.
• Professional CGM use was also associated with a higher proportion of insulin initiation (p<0.0001). However, the proportion of professional CGM users who initiated insulin after using professional CGM was still quite low at 20%. This figure is higher than in the non-CGM-using group (10%). However, the fact that only one in five professional CGM users initiated insulin suggests that despite having the added glycemic information provided by CGM, and despite the fact that none of these patients were achieving an A1c <7% at baseline, providers were still generally not initiating insulin.
• When broken down into those who did and did not initiate insulin, the professional CGM users still saw a greater A1c improvement than their non-professional CGM-using counterparts. Specifically, professional CGM users who initiated insulin saw a -0.6% A1c improvement (8.9% to 8.3%) whereas non-CGM users who initiated insulin saw a slight +0.1% A1c increase (8.8% to 8.9%) (p<0.0001 for between-group difference). Likewise, professional CGM users who didn’t initiate insulin saw a -0.9% A1c improvement (8.7% to 7.8%) compared to a -0.4% A1c improvement in the non-professional CGM group (8.6% to 8.2%) (p<0.0001 for between-group difference). Thus, while the low proportion of insulin initiation after professional CGM use is discouraging, these results are still promising, as they suggest that professional CGM can still result in glycemic benefits even if its use is not associated with a therapeutic change. This again hits on the value of CGM to drive behavioral change, reiterating the findings of the MOBILE and FLASH-UK RCTs, both of which showed that those who initiated CGM saw glycemic benefits despite no change in insulin dosing, regardless of the population studied (MOBILE was type 2s on basal-only; FLASH-UK was type 1s).
• The researchers also explored associations between professional CGM use and non-insulin diabetes medication use. Overall, they found that initiating professional CGM was associated with significant reductions in sulfonylurea and biguanide use and significant increases in GLP-1 and SGLT-2 use. There was a particularly stark decline in sulfonylurea use among those who used professional CGM and no change among those who did not use professional CGM. This discontinuation of use among CGM users could be due to previously unknown hypoglycemia appearing in the CGM tracings, although this is purely conjecture on our part. We were also glad to see that GLP-1 and SGLT-2 use increased among those using professional CGM, although we would have liked to have seen further initiation of these therapies, given this population’s higher A1c values and the medication’s complication-risk-reduction benefits. Use of these classes isn’t an end in itself, of course; dQ&A shows that nearly half of the 1,500 people with diabetes using this class in the US and outside the US still have A1cs >7%.

rt-CGM use among type 2s on insulin therapy (n=36,080, average A1c of 8.3%) found to meet NHS willingness to pay threshold with an incremental cost-effectiveness ratio of £3,684 per QALY

Health economist Mr. Stephane Roze (Vyoo Agency) presented data from a health economics analysis demonstrating that rt-CGM use among people with type 2 diabetes on insulin therapy meets the NHS willingness to pay threshold. Specifically, rt-CGM use among type 2s has an incremental cost-effectiveness ratio (ICER) of £3,684 per quality adjusted life year (QALY) gained compared to SMBG, falling well-below the cutoff of £20,000 per QALY gained. Mr. Roze based his analysis on patients with type 2 diabetes on insulin therapy (n=36,080) who had an average A1c of 8.3%. Using data from both the MOBILE study and from the Kaiser health system between 2014-2019 of rt-CGM use among type 2s – both of these studies were published in JAMA, wow! – Mr. Roze estimated that rt-CGM use was associated with an average A1c reduction of 0.56% compared to SMBG. Mr. Roze also estimated that rt-CGM use was associated with reduced rates of severe hypoglycemia and hyperglycemia/DKA events of 0 per 100 person-years versus 4/100 person years and 2.5/100 person years, respectively for patients using SMBG. Additionally, Mr. Roze calculated the yearly cost of rt-CGM use at £1,250 (36 sensors per year assuming a 10-day sensor life at the current price of Dexcom G6 in the UK – based on these figures the average cost to the NHS per G6 sensor appears to be ~£35). For comparison,  the yearly cost of SMBG totaled £402 based on results from the DIAMOND trial, which demonstrated an average of 3.8 BGM tests/day among patients with type 2 diabetes on insulin therapy. Together, these estimates were used to develop a model for rt-CGM cost effectiveness, which was assessed via the IQVIA CORE Diabetes Model. Based on this analysis, Mr. Roze calculated total direct costs related to diabetes of £79,866 for rt-CGM users and £77,172 for SMBG users resulting in a difference in cost of £2,694. In this analysis, the  additional upfront cost of rt-CGM was largely offset by an expected reduction in complications (DKA, severe hypoglycemia, retinopathy, amputations, nephropathy, and cardiovascular disease) for people using rt-CGM compared to SMBG resulting in the relatively close total direct costs of diabetes for both cohorts. Specifically, in 71% of simulations run based on Mr. Roze’s model, rt-CGM use among people with type 2 diabetes on insulin met the willingness-to-pay threshold of costing <£20,000 per QALY gained. Furthermore, in 39% of simulations run based on Mr. Roze’s model, rt-CGM went beyond cost-effectiveness and had the potential to drive cost-savings with a willingness to pay threshold of <£20,000 per QALY gained. Of note, this analysis only included direct costs associated with a diagnosis of type 2 diabetes and did not factor in potential costs from lost productivity, which Mr. Roze argued means real-world cost-efficacy of rt-CGM is likely even higher than demonstrated with this data. We see this analysis as encouraging in terms of building support for CGM use among patients with type 2 diabetes on insulin therapy, which was a key component of updates in the ADA 2022 Standards of Care.

• Updated NICE guidelines published in March recommend offering is-CGM to adults with type 2 diabetes on MDI therapy citing the higher cost of rt-CGM as a reason for not recommending use of the technology in this population. However, the NICE guidelines do note that providers should “consider [rt-CGM] as an alternative to is-CGM for adults with insulin-treated type 2 diabetes if it is available for the same or lower cost.” With the UK launch of FreeStyle Libre 3, which brings rt-CGM functionalities to the FreeStyle Libre franchise CGMs as the same price point as the is-CGM FreeStyle Libre 2 model plus this novel analysis demonstrating cost-efficacy for rt-CGM using Dexcom G6 as the base case among type 2s on insulin, we are curious if NICE may provide further guideline updates or if UK providers will begin to offer rt-CGM to more of their patients with type 2 diabetes.

Observational study from UK NHS audit (n=14,248) shows FreeStyle Libre use correlated with significant decrease in prevalence of impaired hypo awareness; fourfold relative risk of severe hypo (!) observed for those with impaired hypo awareness vs. those with normal awareness

During a midmorning oral presentation session, Dr. Beatrice Pieri (King’s College London) presented data from the Association of British Clinical Diabetologists’ (ABDC) audit showing FreeStyle Libre use can substantially reverse impaired hypoglycemia awareness (IHA). This multicenter, observational study (n=14,248, mean age 43, 96% people with type 1 diabetes, 50% female) consisted of data collected during the ABCD audit between November 2017 and August 2021. Dr. Pieri explained that the investigators used the Gold Score as a primary outcome measure, as it is a validated screening tool to assess hypoglycemia unawareness (score ≥4 indicates impaired hypoglycemia awareness and 7 indicates complete loss of awareness), as well as the DDS2 two-item diabetes distress screening tool (score ≥3 indicates moderate distress and <3 indicates low distress). At baseline, mean Gold Scores and DDS2 scores across the entire cohort were 2.7 and 3, respectively; though these are favorable, nearly 30% (n=3,477) of participants had impaired hypoglycemia awareness, indicating major risk of severe hypoglycemia and challenging follow-ups.

• Of the 4,391 participants with a paired baseline and follow-up Gold Score, FreeStyle Libre use was associated with a significant decrease in the prevalence of IAH, from 28% to 18%. This decrease was also accompanied with a decrease in the prevalence of severe hypoglycemia (from 14% to 5%) as well as in the prevalence of total loss of hypoglycemia awareness (from 3.7% to 3.2%). Notably, the number of severe hypoglycemia episodes decreased from 140 per month before CGM use to only 48 per month after CGM use, and most impressively, 55% of people with impaired hypoglycemia awareness at baseline (n=673 of 1,233) had restoration of hypoglycemia awareness, as evidenced by a Gold Score <4. The potential for less hypoglycemia is significant.
• Among people with IHA, the prevalence of those experiencing ≥one severe hypoglycemia episode was 37%, compared to only 11% of people with normal awareness (p<0.001). Concerningly, this translates to nearly a fourfold increase in risk of severe hypoglycemia in those who had impaired hypoglycemia awareness. This finding certainly reinforces the importance of CGM as a tool to help people understand when they are at risk for hypoglycemia and take preventative action. Especially in light of the preliminary results from the Hypo-RESOLVE study presented at EASD 2021 showing a disturbing increased risk in microvascular and macrovascular complications following any hypoglycemia exposure, this study’s results speak to the importance of giving people with diabetes the tools that they need to best manage their glycemic levels.
• Factors associated with increased likelihood of having impaired hypoglycemia awareness at baseline were: (i) increased diabetes duration (OR=1.02, p<0.001); (ii) older age (OR=1.01, p<0.001); (iii) higher diabetes distress (OR=1.36, p<0.001); and (iv) number of severe hypoglycemia episodes (OR=1.31, p<0.001). After using FreeStyle Libre, a shorter duration of diabetes and a higher Time in Range were positively associated with restoration of hypoglycemia awareness.

First analysis out of Dexcom’s Type 2 Help study: Three-month observational study suggests Dexcom G6 improves quality of life but doesn’t lead to Time in Range benefits in broad type 2 (n=180) and prediabetes (n=29) population

In the e-poster hall, a 12-week observational study showed that Dexcom G6 improved quality of life in 209 adults with type 2 diabetes (T2D) and prediabetes in the US. The Dexcom-sponsored study, first-authored by Dr. Margaret Crawford (Dexcom), included 209 adults with T2D or prediabetes in the US (average age 56), including 29 people with prediabetes/at high risk for diabetes, 104 not on insulin, 40 on basal-only therapy, and 36 on fast-acting insulin. The study population was more reflective of the general T2D population than many other studies in diabetes technology with 15% of participants identifying as Black, 9% as Asian, and 32% as Hispanic, an effort that was intentional in the study design. The analysis evaluated both Time in Range and quality-of-life/treatment satisfaction scores at baseline and after 12 weeks of use. Surprisingly, none of the subgroups saw an improvement in average Time in Range after twelve weeks, although the Dexcom researchers attributed this to the short duration and the observational nature of the study (didn’t include a directed intervention), and the groups all were achieving or nearly achieving a Time in Range ≥70%. Regardless of the lack of significant improvement in Time in Range, all subgroups saw improvements in their Patient Health Questionnaire-2 (PHQ-2) score, a measure of depressive symptoms on a 0-6 scale with a score ≥3 indicative of major depressive disorder, which in the study was used as a proxy for quality of life. Across all groups, PHQ-2 scores fell significantly to ≤1 from a baseline of ~1-1.5. Notably, those in the high-risk/prediabetes group saw additional significant improvements in self-reported self-efficacy, illness perception, and sleep quality. Likewise, those with T2D not on insulin saw significant improvements in self-efficacy and illness perception. Unsurprisingly, CGM satisfaction scores were high across all groups at ~4.0 on a 0-5 scale – this is unsurprising since we believe that virtually all PWD prefer not to do fingersticks rather than to do fingersticks. We believe value due to other components of CGM is also high, we look forward to seeing more specifics. Overall, the researchers – who included the esteemed Dr. Katharine Barnard-Kelly – argued that the results suggest that there are significant quality of life benefits to be gained by a wide range of people with T2D using rt-CGM.

• This analysis is a part of the broader Type 2 Help study (NCT04503239) and is the first analysis to come out of that study. Based on ClinicalTrials.Gov, the study aims to include 306 people total with T2D and prediabetes and is collecting data on CGM metrics, medication use, food intake, physical activity, sleep, heart rate, five-hour OGTT, and quality of life outcomes. Based on ClinicalTrials.Gov (last updated October 2021), the study was set to complete in February 2022, suggesting that it may be fully complete, although this wasn’t addressed in the poster or associated recording. Additional analysis from the study will be read out at ADA 2022 on the relationship between behavioral changes and glycemic outcomes in this broad type 2 population.

Ascensia symposium: Eversense E3 CE-Marking expected “quite soon”; plans for direct integration between E3 and Ascensia Contour BGM for CGM calibrations; two new Eversense CGM studies

During Ascensia’s sponsored symposium, Dr. Francine Kaufman (Chief Medical Officer, Senseonics) gave exciting data and product pipeline updates for Senseonics’ Eversense CGM franchise. Most notably, on the company’s next-gen Eversense E3 CGM that was approved by the FDA in February 2022 and launched recently in April 2021, Dr. Kaufman shared that on the European timeline, “we know that we will hear quite soon, and we’ll be very excited to come back … to Europe.” As a reminder, while Senseonics has launched an 180-day sensor (Eversense XL) in Europe, that system is adjunctively labeled and requires more calibrations than Eversense E3, and so the CE-Marking of E3 would mark a noteworthy system update for users in the region. It’s worth noting that Dr. Kaufman referred to the CE-Marking of E3 as “imminent” during an interview earlier this month, but also noted that “the word imminent has no meaning to [her] anymore whatsoever” on account of regulatory uncertainties in Europe, leaving us curious as to whether Senseonics has had any favorable updates from the EMA since we last spoke.

• On Senseonics’ pipeline, Dr. Kaufman shared an exciting update that the company will aim for direct CGM connectivity with Ascensia’s Contour BGM systems, which will likely streamline the calibration requirement for Eversense CGMs. It’s great to see Senseonics create synergies through its commercialization agreement with Ascensia, and we imagine that the direct BGM to CGM transmission will greatly simplify the calibration process. Elsewhere in the pipeline, Senseonics reiterated that it is working on an insulin pump integration for sensor augmented pump therapy (e.g., Beta Bionics) and a Gen2 insertion tool for providers who implant Eversense CGMs. Dr. Kaufman also touched on Senseonics’ upcoming CGM devices in great detail, reiterating her points made at ADCES 2021.

 

• Dr. Kaufman and Dr. Federico Boscari (University of Padua) touched on two new Eversense CGM studies, both of which were published in 2022. In a randomized crossover trial where participants (n=16) used either Dexcom G5 or the implantable Eversense system, the G5 was slightly less accurate vs. BGM (MARD=13.1%) compared to Eversense vs. BGM (MARD=12.3%, p<0.001). More notable, however, was a table comparing participant opinions about Dexcom G5 vs. Eversense, where it appears as if users liked Eversense better overall. Broken down further, the participants preferred Eversense’s associated app, portability, alarm utility over G5 (see the first figure below). In the second study, which was a randomized, prospective, national, multicenter study (n=149), the Eversense CGM system led to significant decreases in Time Below Range for patients who are prone to hypoglycemia (see the second image below).

• During an introductory presentation, Dr. Moshe Phillip (Schneider Children’s Medical Center of Israel) argued that “CGM is the archimedean point of diabetes technology.” In lay terms, Dr. Phillip was explaining how the latest advancements in diabetes technology following the advance of CGM (e.g., insulin pumps, AID, next-gen AID, decision support systems, etc.) have only been made possible as a result of CGM data. Dr. Phillip spent the rest of his presentation highlighting the role of decision support as paving the way for the future of diabetes technology, saying that these systems can help both HCPs and people with diabetes receive a suggestion based on an amalgamation of expertise from all the researchers, clinicians, companies, and datasets in the world. His presentation largely aligned with his most memorable session from EASD 2021, where he and Dr. Tadej Battelino (University Medical Center Ljubljana) highlighted improved outcomes from AID systems and the need for decision support at the patient and provider levels.

Massive Dexcom G6 real-world studies: Greater engagement with G6 features associated with +54 minutes/day in Range vs. low engagement (64% vs. 60%, n=46,633); US study (n=69,375) shows G6 feature engagement correlates with likelihood of meeting 70% Time in Range target

Kicking off day #3’s Dexcom-sponsored symposium, Dr. Giada Acciaroli (Staff Data Scientist, Dexcom) presented data from a massive sample of Dexcom G6 users (n=46,633) showing that high engagement with G6 CGM features was broadly correlated with +54 minutes/day in Range vs. low engagement (64% vs. 60%, respectively). In this real-world international study sample, Dr. Acciaroli explained that participants were enrolled from Canada (n=7,976), the UK (n=7,473), Spain (n=1,016), the Nordic region (n=9,023), the DACH region (n=16,701, Germany, Austria, Switzerland), Italy (n=3,839), and South Korea (n=605) who started using Dexcom G6 at least 30 days before October 2021. The study lasted six months and included participants who had ≥70% of sensor wear time, were between ages two and 100, and who did not change their alert settings over the course of the study. The investigators created a composite engagement index that reflected user interactions with Dexcom G6 alerts, Dexcom Clarity, and Dexcom Follow (0=lowest engagement, 5=highest engagement). Engagement was similar across diabetes type (average for type 1s=3.95, average for type 2s=3.87). Unsurprisingly, users in the highest engagement quartile (scores ~>4) achieved a +54 minute/day increase in Time in Range (+4%) compared to the lowest engagement quartile (scores ~<2), and while this may not constitute “clinically significant” glycemic improvement according to 2019 consensus targets, we think it’s fascinating to see the heterogeneity of Time in Range outcomes among users who are presumably all already benefitting from using any CGM use altogether.

• Dr. Acciaroli presented a graph (see below) showing the overall use of Dexcom G6 features across age groups. We are pleased to see across-the-board strong use of Dexcom G6 alerts and Dexcom Clarity, the latter of which is how users can assess important glycemic metrics including Time in Range. Interestingly, there was a bimodal distribution of Dexcom Follow use with pediatric and older populations seeing the highest uptake, presumably because these groups have the highest percentage of family, friends, and caregivers who are facilitating diabetes management. Dr. Acciaroli also shared the use of G6 features broken down by self-declared diabetes type, which we’ve included in a table below.

	Type 1s	Type 2s
High alert enabled	78%	73%
Low alert enabled	90%	90%
Urgent low alert enabled	82%	85%
Uses Dexcom Clarity	92%	98%
Uses Dexcom Follow	53%	41%

• Dr. Acciaroli also presented data on the association (or lack thereof) between glycemic outcomes and utilization of individual Dexcom G6 features.
  
  • Alert utilization. In the DACH region (Germany, Austria, and Switzerland), utilization of the high alert (78% of cohort) drove a 1.4 hour/day reduction in Time Above Range (from ~35% to ~30%, p<0.001) relative to those who had it disabled (22% of cohort). Although these results are specific to DACH countries, Dexcom noted that “similar results [were] observed worldwide.” Utilization of the low alert (89% of cohort) drove a 14 minute/day reduction in Time Below Range (from ~1% to ~0%, p<0.001) relative to those who had it disabled (11% of cohort). And last, utilization of the urgent low soon alert (79% of cohort) drove a 4 minute/day reduction in time <54 mg/dL (from ~0.4% to ~0%, p<0.001) relative to those who had it disabled (21% of cohort). Dr. Acciaroli also included a graph showing the alert utilization across the international sample:

• • Dexcom Clarity utilization. In Italy, Dexcom Clarity users saw an additional 1.5 hours/day Time in Range compared to CGM-only users (p<0.001). Similar results were observed worldwide. Dr. Acciaroli also provided a map showing Clarity utilization across the entire sample:

• • Dexcom Follow Utilization. Canadian pediatric G6 users (n=317) who had a caregiver tracking them via Dexcom Follow spent +2.6 hours/day (+11%) in Range compared to those who did not use Dexcom Follow (61% vs. 50%, respectively). Similar results were observed worldwide. Dr. Acciaroli also provided a map showing Follow utilization across several countries in the sample:

• Separately, Mr. Robert Dowd (Senior Data Scientist, Dexcom) presented similar data to Dr. Acciaroli from US Dexcom G6 CGM users (n=69,375), showing that increased G6 feature utilization was associated with a significantly greater percentage of users achieving a 70% Time in Range. This retrospective observational study of 61,299 type 1s and 8,076 type 2s was structured nearly identically to the international tracker, except that it was conducted between June and November 2021. Users were stratified into different “engagement types,” ranging from those who only used CGM at the low end, to people who either subscribed to Dexcom Clarity push notifications or logged into the Clarity app notifications in the middle, to people who both got push notifications and logged into the Clarity app at the high end. The distribution of these engagement types was relatively similar across type 1s and 2s, with ~45%-50% of people in the intermediate category, and roughly 20%-30% of people each in the low and high engagement group. With higher engagement types more likely to meet the ADA-recommended guidelines of 70% Time in Range, we think that there is a huge opportunity for clinicians to further optimize the ways in which their patients are using their CGM, to ensure that people can achieve the best outcomes possible.

• • Much like the international study, greater engagement was correlated with a greater Time in Range. For type 1s, Time in Range was +1.7 hours/day higher in the high vs. low engagement groups (65% vs. 58%), and for type 2s it was +1.6 hours/day higher (67% vs. 60%).

• • In the US, utilization of Dexcom Clarity was high among both type 1s and 2s, sitting at 74% and 78%, respectively. We continue to think that Dexcom Clarity is an essential component of the Dexcom G6 system given that it provides CGM-derived insights including Time in Range.

• Interestingly, many users (both type 1s and 2s) opted not to receive Dexcom Clarity notifications. We imagine this is because repeated notifications can contribute to alarm fatigue. However, we also know that the insights from the Clarity app can be powerful tools for people to assess their diabetes management. We hope that the new Dexcom G7 app, which integrates the insights of Clarity directly into the glucose reader app, might make CGM-derived metrics more accessible in a way that is less burdensome.

Dexcom ONE real-world data (n=1,859): users spend +1.6 hours/day in Range (66% vs. 59%) after first ten days of sensor use; only +1.2 hours/day in Range at day 30 vs. day 1; alert usage associated with greater Time in Range; affordable CGM option drives greater CGM coverage in Eastern Europe

Dr. Žydrūnė Visockienė (Vilnius University, Lithuania) presented a first glance at real-world data among Dexcom ONE users (n= 1,859) in Lithuania, Latvia, Estonia, and Bulgaria. As background, Dexcom ONE maintains the same form factor as Dexcom G6 but is built on a novel and “simplified” software with a “simplified alarm scheme” (much like Dexcom G7). Dexcom ONE has been available in Bulgaria, Estonia, Latvia, and Lithuania since September 2021 via Dexcom’s e-commerce platform, is not compatible with AID systems, and also does not support remote monitoring or data-sharing functions that have allowed caregivers to remotely track a patient’s glucose levels. Recently, Dexcom announced that it will also bring ONE to the UK and Spain. Turning to the study, Dr. Visockienė explained that participants (n=1,859) uploaded at least one Dexcom ONE CGM value between the product launch and February 2022, meaning that the study data spans five months. Most participants were from Estonia (n=803, 43%) followed by Bulgaria (n=452, 24%) and Lithuania (n=435, 23%), and then Latvia (n=164, 9%).

• During participants’ first Dexcom ONE sensor session, users recorded a 1.6 hour per day improvement in Time in Range (+7% from roughly 59% to 66%). From our view, this result is encouraging but unsurprising given that CGM has the power to help individuals understand when they are hypoglycemic and hyperglycemic and can alert users to correct as needed. We’d be interested in seeing the Time in Range data stratified across whether patients were CGM-naïve or not. At day 30, users (n=923) experienced a sustained 1.2 hour/day Time in Range improvement (5%) from 59% at day one to roughly 64% (p<0.001). We of course would have liked to see data showing that CGM drove improvements in this cohort that resulted in a mean Time in Range above 70% in alignment with international consensus targets. However, considering that Dexcom brought its ONE CGM to countries where CGM penetration was incredibly low among type 1s to begin with, we are encouraged to any improvements whatsoever and hope that access to CGM can drive access to other life changing therapies such as insulin pump therapy, and eventually AID. Additionally, other studies investigating the impact of CGM on Time in Range including MOBILE and FLASH-UK resulted in Time in Range values of 59% and 52%, respectively, suggesting that CGM alone may not be enough to help patients achieve the consensus goal of >70% Time in Range.

• Participants using Dexcom ONE’s high, low, and “delayed high” alerts had a higher Time in Range, on average, than non-users. Users of the “high” glucose alert (93% of cohort) spent +1.7 hours/day in Range compared to those who disabled the alert. Similarly, users who used the “low” glucose alert (95% of cohort) spent +29 minutes/day in range compared to those who disabled it – we may be a minority, but from our view, only those with major problems (not, like, just wanting to turn it down if they go to an opera, etc.) should disable it since the root cause of severe hypoglycemia should be something that virtually all PWD can address. Dexcom ONE, similarly to Decom G7, also features a “delayed high” alert that only notifies users when hyperglycemia is detected to be persistent, although only 7% of the cohort elected to use this functionality. While we wonder how many really understood this, we note that  participants who didspent +43 minutes/day in range vs. the 93% of the cohort that disabled the alert. Okay, that’s five hours a week, 260 hours a year, or over ten more days annually time in range!
• Musing on potential areas of improvement for the Dexcom ONE system, Dr. Visockienė noted that Dexcom should allow ONE users to use the Dexcom Follow (iOS, Android) app, which enables CGM data sharing with friends and family. On this, we sincerely agree, and while some say they are curious to better understand the reasons behind Dexcom’s choice to not include this feature for its ONE CGM users, we simply assume that this will be something that makes the G7 differentiated enough to opt for, despite a higher price. Some may perceive  Dexcom’s rationale for excluding certain features from its ONE system as an attempt to ensure that the system is as simple to use as possible, but we imagine that the company has  likely simply opted to save certain “desirable” features for its Dexcom G6 CGM so that it could price that CGM higher. While that might be a mischaracterization on our part, the Dexcom Follow app is clearly an extraordinarily important tool for caregivers and loved ones to help care for people with diabetes and detect emergencies such as hypoglycemia and DKA or even simply arrows moving down rather than sideways. Wee definitely share Dr. Visockienė’s wishes – from our view, this is likely an experiment on Dexcom’s part and we do like it that they have a lower cost option even if it is not the “best” CGM option from Dexcom. Dexcom ONE may be the highest-value option – that’s hard to tell overall..
• Notably (w0w!, Dr. Visockienė noted that the legal availability of an affordable rt-CGM (Dexcom ONE) in Lithuania, Estona, and Bulgaria sped up negotiations with payers in those countries. Now, Dr. Visockienė explained that there has been a “huge change in reimbursement” for type 1s, with CGM becoming accessible to far more people beyond those on insulin pump therapy. Dr. Visockienė explained that a huge reason for this was because of the (relatively) cheaper price of Dexcom ONE compared to other CGMs available in the regions where ONE is sold. See below for a competitive landscape:

Six-month Welldoc BlueStar and Dexcom G6 pilot study in type 2s with baseline A1c of 9.5%: continuous CGM use (n=37) drives +7.4 hours/day in Range (+31%), while “intermittent” wear of CGM (n=55) drives +4.8 hours/day (+20%); ePoster shows continuous CGM use drives +8.9 hours/day (!) in Range from 21% to 58%

Closing out a Dexcom-G6-focused symposium, Welldoc Chief Strategy Officer Dr. Anand Iyer presented data from a pilot study of Dexcom G6 and Welldoc BlueStar program users (n=92). For background, Welldoc’s BlueStar, which has secured nine 510(k) clearances, offers AI-driven coaching, basal titration support, and data integration from BGMs, CGMs, pharmacies, labs, and activity trackers. We last heard from Dr. Iyer at DTM 2021, where he showed how BlueStar can centralize health data from several different streams, including CGM data from Dexcom’s real-time Partner Web APIs. In this six-month pilot study, 92 participants with type 2 diabetes and an A1c ≥8 (50% on orals/non-insulin injectables, <10% on bolus insulin, 44% female) were enrolled across three centers. The primary aim of the study was to study the degree to which the synergistic power of Dexcom G6’s glucose insights and BlueStar’s AI-driven platform could improve glycemic outcomes. The study involved two arms: one with participants who wore Dexcom G6 continuously for the entirety of the six-month period (n=37), and another with people who only wore CGM “intermittently” (n=55). While the study primarily included people who were ages 40-54 (44%) and 55-64 (36%), there were also people ages 18-39 (13%) and 65+ (7%). At baseline, the overall cohort had an A1c of 9.5%.

• BlueStar users who wore CGM continuously had a 2.6 hours/day greater improvement in Time in Range (+11%) compared to those who wore G6 “intermittently.” While we were disappointed to not see baseline or final Time in Range values, Dr. Iyer did say that people who continuously wore CGM saw a +7.4 hour/day improvement in Time in Range (+31%, p<0.01), and those who wore CGM “intermittently” saw a +4.8 hour/day improvement in Time in Range (+20%, p<0.01) over the six-month study. Also notable, people who wore CGM continuously saw a 16 mg/dL greater mean glucose reduction compared to those who used CGM “intermittently” (-52 mg/dL vs. -36 mg/dL, both p<0.01). There was no statistically significant change in percent CV.
• Welldoc also presented an ePoster (EP170) with similar data in type 2s whose baseline glucose was ≥180 mg/dL (n=39). Those with “highest” CGM use (n=11) spent +5.8 hours/day in Range vs. those with “lowest” use (n=12). In the highest CGM group (defined as continuous use over 24 weeks), Time in Range improved from +8.9 hours/day from 21% to 58% (p=0.0037). Those with “intermediate” CGM use (defined as use over 13-<24 weeks) improved +7.0 hours/day from 18% to 47% (p=0.0009), and those with “lowest” use (defined as use over <13 weeks) improved +3.1 hours/day from 21% to 34% (p=0.04). Time Below Range did not significantly change in either arm, and while GMI improved in the “highest” use arm (from 9.1% to 7.6%, p=0.0025) and in the “intermediate” arm (from 9.2% to 8.2%, p=0.02), the “lowest” arm saw no statistically significant change in GMI.

Stanford’s “4T” pilot study improves outcomes for new-onset pediatric type 1s using early CGM initiation and timely interventions: percentage of people with A1c <7% one year after diagnosis improved from 28% to 53%

Dr. David Maahs (Stanford University) presented positive results from the “4T” pilot study of newly diagnosed pediatric type 1s, demonstrating a significant improvement in glycemic outcomes using the “4T” approach compared to historical data. The four Ts in the “4T” pilot study name are teamwork, targets, technology, and tight control. The goals of the 4T pilot are succinctly outlined in the figure below. When reviewing historical new-onset pediatric type 1s, A1c starts out the highest at time of diagnosis, drops to a nadir around months 5-6 after diagnosis, and then starts to creep back up. The aim of the 4T pilot is to help patients reach that nadir and keep A1c at that lower level.

• The 4T intervention centers around earlier CGM initiation and timely interventions. Under the 4T framework, patients were initiated on CGM just 1-14 days after their diagnosis with type 1 diabetes. This CGM initiation with a CDCES lasted 2-3 hours for training and was followed by a second 2-3-hour virtual visit with a nurse practitioner a week later for check-in. Following CGM initiation, these patients continued on the same visit schedule as was previously used at Stanford, with the addition of supplemental telehealth visits, as appropriate based on CGM data.

• The majority of patients in the 4T pilot opted into remote monitoring of CGM data. This allowed Stanford engineers to build out a “Timely Intervention for Diabetes Excellence” (TIDE) tool. The TIDE tool (an earlier version is described in a 2021 JMIR article) takes CGM data from these patients, analyzes the data, and presents results in a dashboard that makes it easy to identify patients that may need more or less help. Despite these metrics being fairly simple (e.g., Time in Range, time in hypoglycemia, etc.), having all of the information in a quickly searchable and filterable dashboard was helpful for 4T’s healthcare providers to offer timelier interventions.

• The pilot 4T study was rolled out from 2018 to 2020 and results were compared against historical data from 2014-2016. A1c results are shown in the plots below, with the 4T data shown in blue and historical control data shown in black. At time of diagnosis, the 4T group started with a much higher A1c at 11.5%, compared to 10.2% for the historical comparator. Both groups reached a similar nadir between months 3-6 after diagnosis; however, the mean A1c of the historical group creeped up over time, while the 4T group stayed much flatter. By month twelve following diagnosis, the 4T group had a mean A1c of 7.4% compared to 7.9% for the historical group. Additionally, the A1c at month 12 for the subset of 4T participants who utilized remote monitoring was even lower at 7.3%.

• Similarly, the 4T pilot group saw about half (49%) of participants achieve an A1c below 7% at fifteen months following diagnosis, more than doubled the rate of the control group at the same time point (21%). In the control, a majority of patients had an A1c below 7% three months following diagnosis; however, this figure drops off quickly to 28% one year after diagnosis and 21% at fifteen months. By comparison, in the 4T group, 63% of participants achieved an A1c below 7% three months following diagnosis and while a drop-off was seen, the drop-off was more gradual. At one year after diagnosis 53% of participants still had an A1c below 7%.

• Looking ahead, data from the full study of 4T will be presented at ADA 2022. The full study includes participants enrolled between July 2020 and April 2022 with all participants receiving remote monitoring. Dr. Maahs gave a  preview of this data which looks even more impressive than the results from the 4T pilot study. In the plot below, the blue line is well below the 7% threshold at 12 months, suggesting a significant majority of the participants in the 4T study were able to achieve an A1c below 7% - wow! Though this is our speculation, this could be related to the Stanford care teams having greater experience with the 4T program, as well as the full study targeting a lower A1c of 7% compared to the pilot study’s 7.5%. We are very much looking forward to seeing the full results in June at the ADA Scientific Sessions.

• • Looking even further ahead, a second study with 4T has already begun and includes AID initiation between months 1-3 after diagnosis. Dr. Maahs also noted that the group plans to scale the 4T program to Spanish-speaking families and to share the approach with other clinics. Additionally, there are plans to adapt the 4T approach to people with newly diagnosed type 2 diabetes in the future.

Landlines in the Age of Smartphones: Dr. Ramzi Ajjan Reviews the Small, But Growing Evidence Base for the Use of CGM in People with Type 2 Diabetes Not Using Insulin

During Abbott’s afternoon symposium, Dr. Ramzi Ajjan (University of Leeds) gave a well-balanced overview of the evidence base around the use of CGM in people with type 2 diabetes not using insulin. While the evidence is still fairly limited and other challenges to widespread CGM utilization still exist (cost and education, to name a few), ultimately, Dr. Ajjan memorably compared using fingersticks for people with type 2 diabetes and CGM for people with type 1 diabetes to giving one group of people landlines and another group of people smartphones.

• Dr. Ajjan began his presentation with a brief discussion on the use of flash glucose monitoring in the UK, a topic that was also touched on by Dr. Pratik Choudhary (Leicester Diabetes Centre) in his preceding talk. FreeStyle Libre first became available for users to self-fund in 2015, but did not become fully available to people with type 1 diabetes until 2020 following years of advocacy work led by Diabetes UK (e.g., the “Fight for Flash” campaign). As of September 2021, FreeStyle Libre use by people with type 1 diabetes is 50% compared to just 11% in 2019. Despite these successes for people with type 1 diabetes, the NICE guidelines currently recommend offering flash glucose monitoring to adults with type 2 diabetes on MDI if they have significant risk for hypoglycemia or have a condition that prevents them from taking fingersticks.

• Before jumping into the heart of his talk, Dr. Ajjan also highlighted a poster presented at this year’s ATTD that confirmed the effectiveness of flash glucose monitoring in people with type 1 or type 2 diabetes. While meta-analysis of 75 trials was not limited to people with type 2 diabetes not on insulin, the results were very conclusive. In the 28,107 people with type 1 diabetes included in the studies, mean A1c reduction was 0.53%. Similarly, in the 2,415 people with type 2 diabetes included, mean A1c reduction was 0.45%. A meta-regression of these studies also showed that greater baseline A1c was associated with greater improvements using flash glucose monitoring and that these A1c improvements were sustained out to at least 24 months for both type 1s and at least 12 months for type 2s. The poster, authored by University of Cambridge’s Dr. Mark Evans, is the latest in a growing evidence base to support the use of CGM in people with type 2 diabetes (particularly those using insulin).

 

• Honing in on the evidence base around people with type 2 diabetes not using insulin, Dr. Ajjan highlighted three studies: Wada et al. (2020), LIBERATES (2020), and Wright et al. (2021). In Dr. Wada’s study, mean A1c was reduced from 7.8% to 7.4% in the flash glucose monitoring group (n=49), while mean A1c was reduced from 7.8% to 7.5% in the SMBG group (n=51) after twelve weeks; interestingly, twelve weeks after the devices were taken from the subject, the difference was more pronounced with the previously-using-CGM group at 7.4% and the SMBG group at 7.7%. In the LIBERATES trial, use of flash glucose monitoring was not associated with A1c improvements, but was associated with 28 fewer minutes per day in hypoglycemia. Finally, in Dr. Wright’s observational study, initiation of flash glucose monitoring was associated with a significant A1c reduction from 10.1% to 8.6% after a mean of ~two months.

• With the evidence base around use of CGM in this population of non-insulin users still limited, we found Dr. Ajjan’s analogy between smartphones and landlines to be particularly apt. The core functionality of a smartphone (i.e., being able to connect to a cellular network) may only be of interest to a small group of people. However, the ecosystem built around smartphones (i.e., apps) that utilize this core functionality make the devices appealing to a much broader audience. Similarly, the core functionality of a CGM – being able to view glucose readings in real-time – may only be valuable to a smaller group of people. However, we are already seeing a much broader CGM-powered ecosystem that allows for more personalized clinic visits, automated insulin delivery, decision support, automated or virtual lifestyle coaching, fitness tracking, and much more that is ultimately opening up this technology to a much broader group of people.

PROs from MOBILE study demonstrate increase in “openness” measure of feeling freer from constraints of diabetes management among participants in CGM arm compared to BGM at eight months

Dr. David Price (VP Medical Affairs, Dexcom) presented patient reported outcomes from the landmark MOBILE study that was read out last year at ATTD 2021 and simultaneously published in JAMA. As a reminder, MOBILE enrolled 165 people with type 2 diabetes who were randomized to initiate CGM (n=165) while the control group (n=57) used BGM. At baseline these two groups had average participant ages of 56 and 58 years old, respectively, and the majority of participants identified as racial or ethnic minorities, did not have a college degree, and were on public insurance, which Dr. Price explained is relatively representative of the US population of people with type 2 diabetes. Turning to patient reported outcomes, participants in the MOBILE study took the Glucose Monitoring Satisfaction Survey (GMSS), developed by behavioral psychology expert Dr. William Polonsky, at both baseline and eight months and saw improvements in “openness,” emotional burden, and behavioral burden. Interestingly, there was no significant difference in emotional or behavioral burden between CGM and BGM users at eight months, though both groups did see a in improvement in both metrics across the study period. Discussing the lack of between group difference, Dr. Price hypothesized that this may have been due to the fact that BGM users were still enrolled in a clinical trial during which their data was reviewed by providers and used to adjust diabetes management, likely allowing patients to feel more supported in their diabetes management, which could certainly result in a reduction of both emotional and behavioral burden. Conversely to emotional and behavioral burden, there was a significant difference in the “openness” experienced by participants on CGM versus BGM, with those in the CGM arm feeling significantly more freedom than those on BGM from constant diabetes management after eight months (p=0.003). 

• The GMSS “openness” component assessed patient perspectives on the following four statements: (i) this tool helps me feel more satisfied with how things are doing with my diabetes; (ii) this tool helps me feel less restricted by diabetes; (iii) this tool helps me be more spontaneous in my life; (iv) this tool helps me be more open to new experiences in life. According to Dr. Price, the continued evaluation of patient reported outcomes is a key aspect of research on diabetes technology and Dr. Price also referenced the GMSS developer, Dr. Polonsky, and his view that transitioning toward assessment of “openness” should be incorporated into PRO analysis that has, up until this point, focused largely on diabetes burden. While we see openness and burden as related – with openness reflecting a lack of burden and burden recognizing challenges people with diabetes face – we see this focus on openness as a positive interpretation of data that is more often presented in the negative. Specifically, the focus on openness versus burden is reminiscent of Dr. Diana Isaacs’ DATAA model for diabetes device data interpretation that emphasizes highlighting days and times patients are doing well over times patients may have struggled with glycemic management suing positive reinforcement to help drive behavioral change.

Non-adjunctive inpatient ICU CGM use during COVID-19 enabled by device validation; CGM enabled 72% reduction in POC tests with 0% Time Below Range by Day 2 of sensor wear

Dr. Eileen Faulds (The Ohio State University) presented data from inpatient use of Dexcom G6 during the COVID-19 pandemic demonstrating that CGM was used safely and effectively to help providers manage critically ill patients with diabetes. Specifically, among 50 patients in the medical ICU (92% on ventilatory support, 46% on a vasopressor, 34% on dialysis, and 74% on steroids), sensor MARD ranged from 8.0%-15.3% depending on a number of factors including oxygen saturation, pH, blood pressure, and partial oxygen pressure. Remarkably, Dr. Faulds shared that despite these factors, patients had an average Time in Range while in the ICU of 72% with no recorded hypoglycemia after day #2 of sensor wear. Additionally, Dr. Faulds shared that because nursing staff were able to use CGM non-adjunctively following sensor validation (more on this below), the frequency of point-of-care testing decreased 72% from 24 tests/day prior to CGM use to an average of 10 tests/day on Day 1 of ICU hospitalization and 7 tests/day on Day 2 of ICU hospitalization reducing the amount of PPE needed by nursing staff and reducing the time nurses had to spend collecting blood glucose data. Based on these data, Dr. Faulds explained that CGM allowed nurses to maintain inpatient glycemic management without the need for frequent point-of-care tests, and also shared her view that this is only the beginning of CGM use for inpatient care. As Dr. Faulds expressed, during the COVID-19 pandemic, CGM technology has enabled nurses to get the same necessary glucose data from their patients, but now via a remote technology. However, according to Dr. Faults, this “masks” the potential of CGM to be a revolutionary technology for inpatient management by utilizing information on trends and alerts to better inform care.

• At The Ohio State University, Dr. Faulds worked with nursing leadership, the diabetes consult team, internal medicine providers, and hospital administrators and develop a safe and effective protocol for implementing CGM in the hospital including the use of device validation. More specifically, for hospitalized patients in the ICU on insulin, nurses would apply a G6 sensor and perform one point-of-care test after the sensor’s warm-up period. If the glucose values from both the G6 and the point-of-care test fell within the +/- 20/20 bounds, then the sensor was considered “validated” and the nursing staff could continue to use the sensor non-adjunctively to inform insulin delivery. Nursing staff were asked for validate individual sensors every six hours with a point-of-care test, and if a sensor failed to be validated, nursing staff could wait an hour or two before trying again to validate the sensor and continue with non-adjunctive care. Of note, 67% of sensors were validated based on the first glucose reading they produced, indicating strong early accuracy. To enact this protocol, Dr. Faulds worked with nursing leadership to create a scaffolded education structure wherein a member of the diabetes consult team would teach nursing leadership how to insert and use a CGM as well as interpret the data. Nursing leadership would then place CGMs and train medical ICU staff who would subsequently train each other at any shift changes. While this was the initial education process, Dr. Faulds shared that nurses wanted to take ownership over CGM management for their patients and CGM training is now incorporated into medical ICU annual competencies for nursing staff. We see this as a big win for inpatient CGM use as nursing staff are absolutely critical for managing glycemic levels among hospitalized patients and this level of buy-in is encouraging and absolutely necessary for the sustained use of CGM. Indeed, we await glucose as the sixth vital sign and hope the field doesn’t have to wait too long.
• Discussing the benefits and limitations of inpatient CGM use, Dr. Faulds highlighted the value of a continuous data stream, but recognized there have been some concerns around accuracy. Starting with the potential limitations of inpatient CGM use, Dr. Faulds noted: (i) outstanding questions on the accuracy of CGM among hospitalized patients as well as potential interferents; (ii) the lag time between interstitial and blood glucose levels and any potential treatment implications; (iii) staff unfamiliarity with the technology; (iv) lack of EHR integration; and (v) cost. However, Dr. Faulds also noted numerous benefits of inpatient CGM use including: (i) continuous real-time data stream; (ii) data-enabled predictive capabilities; (iii) threshold and predictive alerts and alarms; (iv) takes less time for hospital staff; and (v) only includes one invasive procedure compared to hourly fingersticks.

Dexcom G6 associated with reductions in the number of inpatient visits over 12 months from 160 to 74 (p<0.001) and 91 to 47 (p<0.001) in type 1s (n=806) and type 2s (n=337), respectively

Dr. Katia Hannah (Dexcom) presented data demonstrating a reduction in hospitalization and length of hospital stay after initiating Dexcom G6 CGM use in people with type 1 and type 2 diabetes. Dr. Hannah presented retrospective observational data from people with type 1 diabetes (n=806) and intensively managed type 2 diabetes (n=337) who initiated Dexcom G6 use investigating hospitalization rates for the 12-months prior to (baseline) and following (follow-up) G6 initiation. Of note, the type 2 population evaluated in this study had an older average age at baseline of 53 years compared to the type 1 population with an average age of 39 years. Additionally, Dr. Hannah described the type 2 population as “more medically complex,” highlighting a higher rate of comorbidities.

Turning to results, first of all, and very excitingly, patients with type 1 diabetes saw a significant reduction in their number of inpatient visits from 160 at baseline to 74 at follow-up (p<0.001). For people with type 2 diabetes, inpatient visits also decreased significantly from 91 at baseline to 47 at follow-up (p<0.001). While there was a reduction in the number of emergency department visits following G6 initiation for both people with type 1 and type 2 diabetes, though this reduction was not statistically significant. While this was an observational and retrospective study, and thus not able to assess any potential causation between G6 use and reduced hospitalizations, we would assume that greater glycemic data available to both patients and providers likely played an important role in reducing hospitalizations. We are also encouraged to see these improvements evident in people with type 1 and type 2 diabetes continuing to demonstrate the benefit of CGM across a wide population of people. Furthermore, we are curious how the cost savings from reduced hospitalizations may compare to the upfront cost of initiating patients on CGM, as we know that cost-savings analyses are often integral for expanded reimbursement and access.

Dr. Guido Freckmann on glucose monitoring in critically ill hospitalized patients: Hybrid point of care-CGM approach may be best due to CGM accuracy concerns, but further investigation necessary

Dr. Guido Freckmann (Institute for Diabetes-Technology, Ulm, Germany) presented on the accuracy of in-patient glucose sensors in people with acute or chronic comorbidities. As the push to bring CGMs into hospital settings progresses, it’s particularly important to understand how diabetes comorbidities and medications impact sensor accuracy, as this has both regulatory and clinical implications. While these are concerns for general CGM use as well, it is of particular importance in the hospital, which disproportionately sees those with comorbidities. Reflecting on recent research, Dr. Freckmann summarized that although CGM accuracy has improved over the years and testing in-hospital settings has shown that CGM is generally accurate, CGM still may not be as accurate in patients with comorbidities or when taking therapies or using technologies that are known to cause interference. This is logical, in our view, and not a surprising result though an interesting one given the high interest in inpatients. For example, a recent pilot study measuring the accuracy of CGM (Dexcom G6) after cardiac surgery in patients with nephropathy found that the MARD of patients with an eGFR>20 mL/min/1.73m² was 12.1% compared to a MARD of 21.3% for patients with an eGFR <20 mL/min/1.73m². Moreover, data published in 2021 found that the MARD of a FreeStyle Libre device increased from just under 22% to nearly 26%  after dialysis. Dr. Freckmann also highlighted data showing that the accuracy of CGM is decreased during events of hypoglycemia and severe anemia but can still serve as a reliable tool to treat hospitalized patients. While there are some studies on the accuracy of in-patient CGM use in critically ill-patients, further studies evaluating standardized procedures and metrics are likely needed before CGM becomes the standard of care. Ultimately, Dr. Freckmann advocated for a hybrid approach to care, suggesting that CGM can be used as a supplement to point of care glucose testing in hospitalized patients. While we imagine that this result would have been expected, we think it is good to see data around this. Importantly, providers must consider that, especially in critically ill patients with comorbidities, the CGM results should be interpreted with caution.

• Dr. Freckmann carefully outlined the pros and cons of using CGMs in hospitals, adapted from Galindo et al. 2020 outlining practice guidelines.

Pros	Cons
Full 24-hour glycemic profile and real-time tracking	Lack of regulatory approval – quality control
Predictions of hypo and hyperglycemia	Validation of CGM accuracy is still needed
Alarms	Added burden on nursing to learn new technology, troubleshooting errors, decision making related to CGM data
Less labor-intensive	Time lag
Decrease contact with the patient	Skin related issues may interfere with care
Reduce the risk of infection	Substance interference

“Don’t panic, reflect in real-time”: Dr. Anku Mehta on actionable use of CGM in pregnancy highlighting alert settings, behavior change, and insulin dosing

In a Dexcom-sponsored symposium, Dr. Anku Mehta (West Hertfordshire Teaching Hospitals NHS) discussed the use of Dexcom G6 in pregnancy, highlighting the utility of real-time data to help pregnant people with diabetes make informed treatment and lifestyle decisions. As Dr. Mehta explained, diabetes during pregnancy is becoming more prevalent as the rates of type 2 diabetes, prediabetes, and obesity continue to rise, putting both pregnant people with diabetes and infants at higher risk for complications, including pre-eclampsia, increased risk of infections, progression of diabetes-related complications, large for gestational age (LGA) birth, postnatal hypoglycemia, and increased risk of developing obesity and type 2 diabetes. With these challenges, Dr. Mehta positioned CGM as a key tool to help pregnant people with diabetes improve diabetes management and ultimately maternity outcomes. Specifically, Dr. Mehta provided guidance around when to check CGM values, when to confirm CGM readings with BGM measurements, best practice for CGM alerts, and interpreting CGM data to inform lifestyle and diet changes. Across these topics, Dr. Mehta emphasized the value of using CGM data to help inform decisions urging providers to use a “don’t panic, reflect” approach with their patients, recognizing that glycemic management is challenging and that many pregnant people with diabetes may struggle to meet the tighter pregnancy-specific Time in Range goal of >70% time between 70-140 mg/dL. Within this paradigm, Dr. Mehta focused on the value of education so that pregnant people with diabetes can see their CGM data and respond accordingly with specific plans in place to prevent hypoglycemia, manage hyperglycemia, and adjust behaviors and diet as necessary. Furthermore, Dr. Mehta reminded attendees that even if someone is doing “everything right” they are still likely to have out of range glucose readings and that these times are not reasons to panic or feel guilty, but are expected experiences that can inform future behaviors and management. Dr. Mehta also highlighted the power of CGM to help identify potentially dangerous glycemic trends in pregnant people with diabetes, including unexplained hypoglycemia in the third trimester, which Dr. Mehta said can actually be an early sign of pre-eclampsia that should be immediately addressed by a health care provider.

• Dr. Mehta recommended that pregnant people with diabetes should check CGM values (i) upon waking, (ii) before meals, (ii) one hour after meals, (iii) two hours after meals, and (iv) before bed. If at any point, the individual feels that their CGM values don’t match how they’re feeling (e.g., signs of hypo or hyperglycemia), Dr. Mehta said that they should be encouraged to do a confirmatory fingerstick, especially if the individual is concerned about hypoglycemia and if CGM values aren’t changing 10 minutes after treatment. Regarding checking glucose values around meals, Dr. Mehta discussed how glucose tends to be highest two hours after eating and that by checking postprandial values at both one and two hours, patients can see how their meal boluses are working and discuss any concerns or changes with providers..
• Dr. Mehta provided best practices for reflectively using CGM data to inform insulin dosing, reminding attendees that insulin resistance can change over the course of pregnancy and that bolus timing may need to be adjusted accordingly. For example, Dr. Mehta recommended that for the first 20 weeks of gestation, pregnant people with diabetes should bolus 15-20 minutes ahead of meals whereas after 20 weeks of gestation as insulin resistance increases, they may want to bolus as far in advance as 45 minutes ahead of meals. When patients do experience postprandial hyperglycemia, Dr. Mehta encouraged providers to teach them not to panic and follow their hyperglycemia management plans, including maintaining at least two hours between correction doses to allow for adequate insulin action time and to avoid hypoglycemia.
• Dr. Mehta discussed alert settings, saying that alerts should only be turned on “if it is something the user can safely, and should, do something about straight away” as an effort to ensure patients can avoid dangerous glucose levels while also recognizing the reality of alert fatigue and frustration. More specifically, Dr. Mehta recommended starting with no alerts or only low alerts and slowly adding in hyperglycemia alerts. Dr. Mehta also suggested that some patients may like to set different threshold alerts at different times of day based on activities or to help patients achieve better sleep.
• Dr. Mehta highlighted the power of CGM data to help inform lifestyle choices, including exercise following meals and adjusting the carb content of meals. In this conversation, Dr. Mehta emphasized the role of providers to help patients identify any areas in their CGM data potentially indicative of challenges with carb counting or where a post meal walk could help keep glucose levels in range. Dr. Mehta then continued that when providers are able to help patients identify these time points, they can help equip them with self-management tools to better manage their diabetes during pregnancy and beyond.

GWave non-invasive radio frequency-based glucose monitor starting clinical trials; aggregate data (n=53) demonstrates 96% readings in Zone A compared to venous glucose

Dr. Irl Hirsch (University of Washington) presented a new dataset on the non-invasive GWave glucose monitor developed by Israel-based Hagar where Dr. Hirsch serves as a medical advisor. We first wrote about GWave back in July 2021 – we are still learning more about noninvasive glycemic monitoring and appreciate the opportunity to learn about the area. Since that time, Hagar closed a Series B funding round for $11.7 million in August 2021. As noted a year ago, the GWave system uses radio frequency to measure glucose via a resistor-capacitor model of the skin and underlying blood vessels. This technology is somewhat different from many other non-invasive sensors that use infrared and light-based spectroscopy measuring reflection of light rays off of glucose molecules.

Hagar’s current GWave prototype is a wrist-worn sensor roughly one-third the size of a smartphone (see picture). Dr. Hirsch explained that the company hopes to miniaturize this technology into a “watch-like” device. There are currently a number of non-invasive glucose monitors under development including other wrist-worn systems including GraphWear, Movano, LifePlus, as well as continued chatter about a potential glucose monitor in a next-generation Apple watch via the company’s partnership with Rockley Photonics - this has yet to be confirmed. We look forward to watching GWave move through the different milestones – reliability, ease of use, pricing, accuracy in larger groups, especially should CGM be commercialized for wellness applications outside of diabetes, sensor insertion, etc. and we remain curious if any non-invasive sensors will be able to achieve non-adjunctive labeling and how they may compare to current next-generation CGMs with Dexcom G7, Abbott’s FreeStyle Libre 3, Medtronic 780G, and Senseonics’ Eversense E3 as those all continue to improve.

• Data from Hagar’s initial clinical trial (NCT04658082) of its GWave system (n=5 with 45 data points each) found that following a 75g oral glucose tolerance test, 98% of GWave readings fell in Zone A compared to BGM while 96% of venous glucose comparators fell in Zone A on the Clarke error grid. While this was a small study, this level of accuracy is encouraging and has now been supported by one trial for a total of seven patients. Specifically, Hagar is currently conducting a trial and aims to enroll 250 people with either type 1 or type 2 diabetes to use the GWave sensor. We are glad to see a larger trial coming – many approaches are with smaller numbers of people and we’re wondering how this trial will be powered, what the enrollment criteria will be, etc. We’re hoping it is a trial with multiple different investigators rather than just one hospital so that more people from different geographies can participate.
• Data from the first nine participants enrolled in Hagar’s larger GWave trial followed a similar patter to the company’s initial trial with 89% of capillary glucose comparators falling in Zone A and 100% of venous glucose comparators falling within Zone A on the Clarke error grid. To date, GWave has been assessed in 53 people with 97% of values in Zone A compared to BGM and 96% of values on Zone A compared to venous blood glucose. Again, this is a small sample size, and while GWave’s current form factor may not appeal to all patients, Dexcom’s first one certainly didn’t either! Should the company be able to successfully miniaturize its radio frequency measurement system, we imagine there could well be a place in the market for GWave’s wrist-worn non-invasive sensors. Traditionally, larger companies don’t do extensive tech work like this but we await more news at who will move the company forward, especially given the recent fundraising.
• Notably, as GWave does not use infrared or light-based technology like some other wrist-worn sensors, experts don’t appear to see the potential for sunlight or skin-tone based interferences. This is good news as the potential to reduce system accuracy by interacting with the light receptors in these devices should be lower – stay tuned and we look forward to seeing and hearing more in the months ahead – it’s a big deal to have Dr. Hirsch’s endorsement.

Dr. Pratik Choudhary suggests that patients ignore CGM data for two hours post meal if need to temper postprandial anxiety, lays out best practices for clinicians in a remote care environment, and advocates for patients establishing “micro-routines”

Concluding the morning sessions of ATTD 2022’s Abbott School, Dr. Pratik Choudhary (University Hospitals of Leicester) presented tips for providing diabetes care in a virtual environment. A salient theme of Dr. Choudhary’s presentation was that diabetes management is often viewed as a zero-sum game but that people can still live long, healthy lives while still not having a “perfect” blood glucose all of the time. The key, Dr. Choudhary stressed, is empowering patients to act on their blood glucose data. Dr. Choudhary noted that a traditional cornerstone of diabetes education is to emphasize the “importance of routine” so that patients can establish healthy eating patterns and successful mealtime insulin dosing, but he suggested that this approach can often seem unrealistic to many patients who lead busy and active lives. Humorously chuckling that the “only place to get routine is [in] jail,” Dr. Choudhary introduced the idea of “micro-routines,” whereby people with diabetes do not endlessly scan their is-CGM after eating due to anxieties around postprandial glycemia. Dr. Choudhary said that for the most part, the type of carbohydrate and the timing and amount of a bolus insulin dose largely “locks in” one’s post-meal glycemic excursion, and so instead, he tells patients to develop a micro-routine around eating where they: (i) scan their is-CGM; (ii) calculate a bolus dose; (iii) bolus and wait 10-15 minutes; (iv) eat; and then (v) “forget” their CGM data for the next two hours to minimize postprandial anxiety. Dr. Choudhary noted that this approach has been particularly effective among “high achievers,” who can often generate more work for themselves because of stress around post-meal glycemia (i.e., they might overcorrect and cause hypoglycemia). We appreciate Dr. Choudhary’s focus on micro-routines and on supporting patients while acknowledging the reality that one’s diabetes management will often be an imperfect process that can be made so much easier by acting on data.

• Turning to virtual care during the pandemic, Dr. Choudhary stressed that the “key elements of diabetes care do not change” when delivered virtually vs. face-to-face. To illustrate his point, he introduced a four-prong framework for clinicians to succeed in a virtual environment, which has much in common with in-person management approaches:
  
  • Right therapy. Dr. Choudhary pointed to a publication that he and his colleagues authored in Diabetes Research & Clinical Practice entitled “Delivering evidence-based interventions for type 1 diabetes in the virtual world - A review of UK practice during the SARS-CoV-2 pandemic.” The paper’s figures contain a series of treatment algorithms that describe how structured education, insulin pump therapy, and CGM can be used effectively in type 1 diabetes management to improve outcomes. Dr. Choudhary stressed that these protocols can enable providers to choose the “right therapy” for individuals even when not seeing them face-to-face. Dr. Choudhary also pointed to the UK Diabetes Technology Network’s education page, which contains a series of informational resources about using diabetes technology for clinicians and patients.
  • Right person. Dr. Choudhary stressed that HCPs delivering virtual care must identify patients with the greatest risk in order to ensure that they can intervene appropriately. Sharing the UK ABCD’s (Association of British Clinical Diabetologists) risk guidance stratification, Dr. Choudhary shared how his group stratifies patients into three levels (green, amber, and red) based on a series of criteria.
  • Right place. Dr. Choudhary called on clinicians to stop asking where they would like to see patients and instead ask where patients would like to see them. To Dr. Choudhary, the “right place” means working with patients to figure out their preferences for the exact modality of care that works best for them.
  • Right time. Dr. Choudhary builds on the ABCD’s risk guidance stratification to suggest annual follow-up requirements based on the red, green, or amber designation of a patient (see below).

• Dr. Choudhary briefly touched on the Diabetes:M app’s (App Store; Google Play) bolus calculator, which takes into account insulin on board and CGM trend arrows. Dr. Choudhary noted that there are not many bolus calculators that consider the direction of glucose levels (another example would be Omnipod 5’s bolus calculator), and that by using this data, the app can calculate a potentially more realistic bolus dose suggestion. We understand that calculating insulin bolus doses can be one of the most challenging aspects of managing diabetes on a daily basis, and so we were intrigued to hear about this solution as a potentially more accurate bolus calculator for patients, although we’ve yet to see any data confirming that this solution is truly more accurate.

Harder, Better, Faster, Less-Invasive? France’s PK Vitality, the UK’s Afon Technology, and the USA’s Profusa Share Early Clinical Results on Non-Invasive and Implantable CGM Systems

As usual, ATTD brought us news from the very cutting edge of glucose sensing technology. As we sauntered through this year’s poster hall, three posters caught our attention, all highlighting work on novel CGM technologies.

• Our poster tour of the world started in France, where PK Vitality shared clinical data on four subjects using its microneedle-based CGM system. The company’s system, known as K’Watch, uses microneedles (<1 mm long) placed in the dermis to measure interstitial fluid glucose. K’Watch is described as “minimally invasive” and comes in the form of a patch that is replaced weekly. The company’s feasibility trial (ClinicalTrials.gov) included 35 subjects with type 1 or type 2 diabetes, though only data from the first four subjects is shared in the poster. One half of the poster is dedicated to the insertion process, which shows post-wear images from four subjects immediately, one day after, and 21 days after sensor removal. Though accuracy results were sparse, the poster claims a MARD of 18% when calibrated with a commercial CGM (unclear if this calibration was performed prospectively or retrospectively). 

 

• Our next stop was the UK, where Afon Technology presented its not-yet-named “non-invasive blood glucose monitoring system.” This wrist-worn device applies a microwave signal to the user and analyzes resonance shifts to estimate blood glucose. The poster shares accuracy data from a small study of three people, two without diabetes and one with diabetes. Using retrospective calibration, overall MARD for the subject without diabetes came in at 12.2% over four days.
• Moving back to our home in California, Profusa shared clinical data from sixteen subjects using its Lumee Glucose sensor. The small hydrogel sensor is implanted in subcutaneous tissue and uses fluorescent molecules that produce light proportional to glucose concentration. Similar to Senseonics’ Eversense, a transmitter sits on top of the skin. Profusa’s clinical study involved sixteen subjects using insulin and included five in-clinic visits with veinous blood draws using the Super GL lab analyzers. During the three-month wear, the average MARD for the Lumee sensor was 13.4% using retrospective calibration.

 

Drs. Sean and Tamara Oser on implementing CGM in primary care

Drs. Sean and Tamara Oser (University of Colorado School of Medicine) presented the two complimentary arms of the PREPARE 4 CGM study on implementing CGM in primary care. The PREPARE 4 CGM study was designed to implement CGM in 60 primary care practices via either practice level education via the American Academy of Family Physicians TIPS CGM training or practice level referral to the virCIS CGM educational webinar and patient and provider resources for six months. Practices enrolling in PREPARE 4 CGM had the choice of whether or not to participate in the TIPS CGM or virCIS educational components and the study was designed with the ambitious goal of recruiting 40 clinics to the “learn” (TIPS CGM) arm and 20 clinics to the “refer” (virCIS) arm. However, both Dr. Osers expressed excitement that within three weeks of opening recruitment, 83 practices have been recruited with 74 practices enrolled, 29 of which enrolled in the “refer” arm – wow! “Practices have been especially thirsty for this and thrilled for the opportunity to participate,” notes Dr. Oser, “as they have commonly expressed eagerness to incorporate CGM into primary care, but they weren’t sure where to look or how to start because of CGM’s focus on subspecialty diabetes care until now. Bringing CGM to primary care, where most diabetes care is delivered in the US, is sorely needed to help reach more people with diabetes, regardless of where they get their care.” Of note, these enrolled primary care practices span a wide range of clinic type including those owned by hospitals or health systems, clinicians, academic centers, rural health clinics, and federally qualified health centers. Currently, the PREPARE 4 CGM trial is just underway, with practice kickoffs starting this month. Of note, PREPARE 4 CGM is working with primary care practices in Colorado only, but we are hopeful that learnings from this trial can be expanded to other primary care clinics in the US.

Diabetes in India: Dr. Mithun Bhartia discusses uptake of professional CGM in India

In his virtual presentation “High uptake of CGM and TIR as a metric in type 1 diabetes and type 2 diabetes in India,” Dr. Mithun Bhartia (Apollo Clinic Guwahati, India) discussed uptake of CGM in India. CGM uptake remains relatively low in that country, and as such, there have been increasing efforts to use professional CGM. However, even professional CGM has faced challenges in adoption. Specifically, FreeStyle Libre Pro was launched in India 2015, but there was relatively poor uptake in the first three years due to confusion over estimated A1c and factory calibration. However, professional CGM uptake has increased significantly in the last three years after several initiatives to increase CGM awareness, including the AGP Clinical Academy organized by the India Steering Committee. Dr. Bhartia said that 175,000 FreeStyle Libre Pros were sold in India in 2021; it is unclear how many users this figure represents given the intermittent use of professional CGM. Access to CGM in India is still limited by the high cost of FreeStyle Libre and Libre Pro CGMs and the country’s self-paid healthcare model, but we are hopeful that professional CGM may make the technology somewhat more accessible. We were also excited to hear Dr. Bhartia mention that India-based company Eris Lifesciences is developing a CGM with more competitive pricing that will launch “soon,” which Dr. Bhartia discussed as a potential avenue to improve access and quality of care in India.

Insulin Delivery Highlights

Beta Bionics’ insulin-only iLet RCT pivotal (n=440): -0.5% A1c improvement and +2.6 hour/day TIR gain on insulin-only vs. standard care of any other insulin delivery method (including MDI, pump, AID) + Dexcom G6 for adults and children with type 1 after thirteen weeks

Speaking to a packed room of rapt listeners, Dr. Steven Russell (Harvard Medical School) read out the highly anticipated results of the investigator-initiated insulin-only iLet pivotal trial, which compared the novel system vs. Dexcom G6 and any other insulin delivery method (including MDI, pump, or AID) in children (n=165, ages 6-18) and adults (n=275). At the end of the 13-week trial, those on iLet saw a +2.6 hour/day Time in Range improvement relative to the standard care group when adjusted for baseline (p<0.001), “almost all” of which occurred within the first day or two of wear. Specifically, those on iLet saw their Time in Range improve from 51% at baseline to 65% at thirteen weeks whereas the standard care arm saw only a slight improvement to 54%. A1c improved a significant 0.5% on iLet vs. standard care (p<0.001 for baseline-adjusted mean difference; n=326). Those on iLet saw their A1c fall 0.6% from 7.9% at baseline to 7.3% at 13 weeks, while the standard care arm saw no change from baseline to 13 weeks, maintaining an A1c of 7.7%.

Topline Results, thirteen weeks

	iLet (adults and children)	Standard care	Mean-adjusted between-group difference
Sample size	219	107	--
A1c	-0.6% 7.9% to 7.3%	no change 7.7% to 7.7%	-0.5%
Time in Range	+14%; +3.4 hr/day 51% to 65%	+3%; +43 min/day 51% to 54%	+11%
Time <54 mg/dl	+0.12%; +2 min/day 0.21% to 0.33%	+0.04%; +1 min/day 0.20% to 0.24%	0.00%

Study Design and System Overview

At baseline, 89% of participants were using CGM and about a third were on MDI (34%), a third were on pump (32%, most on CGM), and a third were on an AID system, most of whom were on Control-IQ (35% total, 23% Control-IQ, 8% MiniMed 670G, 4% PLGS). Children were randomized 2:1 to iLet (n=112) or standard care (n=53), which was defined as Dexcom G6 and their usual insulin delivery method, while adults were randomized 2:1:2 to iLet (n=107), standard care (n=54), and iLet with Fiasp (n=114).

We are eager to (hopefully) see sub-analyses comparing the results for subgroups based on the insulin delivery method used at baseline and in the standard care arm throughout the trial. Furthermore, we hope to see additional outcomes read out, including time above range, time below range, and time >250 mg/dL, which were not reported in today’s presentation, as Dr. Russell focused on the first four outcomes of the hierarchy analysis. Regardless, we’re excited to see these results, given that this system has been closely watched for a long time – for a little bit of additional context, see our coverage of Dr. Ed Damiano’s annual updates from Friends for Life in 2022, 2021, 2020, 2019, and 2018. We’ll be interested to see how the FDA views the data and the system once it begins the review process, the timeline of which has not been disclosed. Furthermore we’ll be watching to see what implications this holds for the bihormonal iLet system, the pivotal of which is now underway, with enrollment beginning in December 2021.

Before we dive into the results, here is a reminder on the features of the insulin-only iLet system. The iLet system uses an in-house tubed pump the “size of a credit card,” which houses the algorithm, and a Dexcom G6. The system requires very little user input: (i) only patient weight is needed for initiation (no dosing parameters); (ii) the only setting to adjust is the glucose target (usual, lower, higher), which can be set for different times of day; and (iii) mealtime announcements do not require carb counting (only breakfast/lunch/dinner and categorical meal size [i.e., “more,” “usual for me,” “less,” or “much less”).

Photo from FFL 2019, from left to right: Gen 4 iLet (used in the pivotal study), Gen 3 iLet, Gen 2 iLet, and iPhone X

Finally, we highlight a few differences in the methodology of the pivotal compared to other AID pivotal trials. For one, this pivotal was an RCT like Tandem’s Control-IQ pivotal, whereas Medtronic’s MiniMed 780G and Insulet’s Omnipod 5 pivotal trials were single-arm studies (i.e., before AID vs. after AID). Second, the study compared participants in iLet to participants in the standard care arm using whatever insulin delivery method they used at baseline and included a baseline cohort almost equally split between MDI, pump, and AID. In comparison, the MiniMed 780G pivotal compared MiniMed 780G to sensor-augmented pump (SAP) or MiniMed 670G, the Control-IQ pivotal compared Control-IQ to SAP, and the Omnipod 5 pivotal compared Omnipod 5 to baseline (SAP or MDI). Third, the study included a sample that is far more representative than the usual AID study. Over a fourth of participants were non-Hispanic Black (10%), Hispanic or Latinx (10%), or another non-White race (6%). While still a minority of the sample, the diversity is far greater than in other AID pivotal trials, the populations of which have been almost entirely non-Hispanic White. This racial and ethnic diversity was intentional on the part of the researchers, who aimed for ≥15% participants of minority race/ethnicity. Likewise, the study worked to be more representative of a wider range of A1c values, setting no upper limit for baseline A1c inclusion and aiming for ≤20% A1c <7% (not quite achieved, as 25% of participants had A1cs <7%) and ≥33% with an A1c >8%.

Detailed Results

• Those on iLet saw a +2.6 hour/day Time in Range improvement relative to the standard care group when adjusted for baseline (p<0.001), “almost all” of which occurred within the first day or two of wear. Specifically, those on iLet saw their Time in Range improve from 51% at baseline to 65% at thirteen weeks, whereas the standard care arm saw only a slight improvement to 54%. This Time in Range improvement in the iLet arm was already seen in the first four weeks of the study and was consistently maintained throughout the study. Dr. Russell argued that although the Time in Range at the endpoint is often discussed in evaluating AID systems, he believes that the change in Time in Range is a more apt measure by which to compare studies given the differences in study populations and methodology. He also noted that this analysis included the results for children, who are known to reach a lower Time in Range with other available AID systems as well, thereby lowering the average. While many would certainly agree with this interpretation, some expressed disappointment to see that, on average, participants still did not achieve the 70% Time in Range target, which has been achieved with other AID systems among adults in both pivotal studies and real-world studies, including those who have challenges in managing their diabetes (see more on this below). From our view, these results are impressive and we’ll continue to watch algorithms and how various groups do – again, we state, coming from this baseline is very impressive and we want to see more trials both start at this baseline and start at higher baselines – all of it is great!

 

• • Dr. Russell also broke out the Time in Range outcomes for children, for adults iLet without Fiasp, and for adults on iLet with Fiasp, all of whom saw Time in Range benefits with iLet. Children saw a 2.4 hour/day Time in Range improvement on iLet compared to those on standard care (p<0.001), from ~47% to 60% within the first four weeks, which was maintained out to week 13. The standard care arm’s Time in Range improved only from 48% to 50%. Among adults, those on iLet (without Fiasp) saw a 2.6 hour/day Time in Range improvement relative to those on standard care, consistent with the overall sample, and saw their Time in Range improve from ~56% at baseline to 69% at 13 weeks, an improvement observed within the first day. The standard care arm saw a slight improvement as well from ~53% to 58% at 13 weeks. Those using iLet with Fiasp saw a further improvement in Time in Range, achieving +3.1 hour/day Time in Range relative the control group, and improving from ~54% at baseline to 71% at 13 weeks, an improvement already achieved in the first four weeks of iLet use. Although Dr. Russell cautioned against looking at the raw Time in Range values achieved, we’d note that among adults, these Time in Range outcomes are lower than achieved with other AID systems (more on this below) and for those using iLet without Fiasp, are lower than the consensus target for Time in Range.
  • Although the specific figures were not read out, Dr. Russell did share that the iLet arm was superior to the standard care arm in time >180 mg/dL and time >250 mg/dL. While iLet was not statistically superior in terms of time <70 mg/dL, during Q&A, Dr. Russell noted that there was a nominal 0.1% decline in time <70 mg/dL among iLet users compared to standard care. This result meets expectations based on the pre-pivotal results, which did not find time <70 mg/dL to be superior with iLet.
• There was no significant between-group difference in time <54 mg/dl, which was already quite low at baseline in both groups (median 0.2% of time <54 mg/dl) and was maintained at 13 weeks, although the iLet group saw a nominal nonsignificant increase to 0.3% (p<0.001 for noninferiority). When separated out, adults and children both saw no significant difference in time <54 mg/dl in with iLet vs. standard care: adults saw a nonsignificant 0.02% difference (p=0.33) and children saw a nonsignificant -0.04% difference (p=0.24). For all time <54 mg/dl data read out, Dr. Russell reported the median value rather than the mean – we’d wonder if there would be a significantly different finding were the other reported.
• At 13 weeks, A1c favored on iLet vs. standard care by 0.5% (p<0.001 for baseline-adjusted mean difference; n=326, doesn’t include those on iLet with Fiasp). Specifically, those on iLet saw their A1c fall 0.6% from 7.9% at baseline to 7.3% at 13 weeks, while the standard care arm saw no change from baseline to 13 weeks, maintaining an A1c of 7.7%. The iLet group also saw a narrowing of its A1c distribution curve (shown below), indicating the elimination of high A1c values and a larger improvement for those with higher A1c values at baseline. Those with A1c values >7% on iLet (n=164) reported a 0.7% A1c improvement relative to their counterparts in the standard care arm (n=76) (p<0.001). This subgroup saw their A1c fall from an average 8.3% to 7.5% on iLet, as compared to falling only 0.1% from 8.2% to 8.1% on standard care.

• • The analysis that Dr. Russell presented also broke down the A1c data for children, adults, and the subgroup of adults on iLet with Fiasp, all of whom saw 0.5% relative improvements in A1c on iLet compared to standard care. Specifically, children saw a 0.5% baseline-adjusted relative improvement with iLet compared to standard care (p<0.001). In the pediatric iLet group, A1c fell from 8.1% to 7.5% while the standard care arm saw no change from 7.9% at baseline to 13 weeks. Likewise, adults (excluding those on Fiasp) saw a 0.5% A1c improvement when on iLet vs. standard care (p<0.001), with the iLet group seeing their A1c fall from 7.6% at baseline to 7.1% at 13 weeks while the control group declined only slight from 7.6% to 7.5%. Those on iLet with Fiasp saw a similar 0.5% A1c improvement relative those on standard care and saw their A1c fall from 7.8% to 7.1%.
• Based on the 24-hour mean glucose profile, a majority of the benefit of iLet came at night, as has been seen with other AID systems. Between 2 am and 8 am, the study saw a widening difference in mean glucose between the two groups, growing to a ~30 mg/dL difference at 6 am when those in the standard care arm were still at a mean of  ~170 mg/dL while those on iLet were far lower at ~140 mg/dL. For the vast majority of the day, those on iLet had a lower mean glucose than those on standard care (exception of around 2 pm-4 pm); however, the between-group difference was far lower. Overall, the iLet group also saw both a lower mean glucose and a tightening of the variability of mean glucose over the 24-hour period relative to the standard care arm, suggesting that those on iLet saw more consistent and lower glucose levels. These results were similar in both children and adults.

• The rate of severe hypoglycemia episodes was higher in the iLet group than in the control group, but not statistically significant (excluding Fiasp users; 18 vs. 11 events per 100 person-years; p=0.39). On an absolute basis, 10 severe hypoglycemic events occurred in the iLet group, while 3 events occurred in the standard care group (as a reminder, the iLet sample was about twice as large). There were no DKA events. Despite the same 2:1 randomization, the group on iLet with Fiasp did not see significantly more severe hypoglycemia events (3 vs. 2) and had a non-significantly lower event rate when normalized by person-years (10 vs. 14 events per 100 person-years; p=0.83). There were two DKA events in the iLet with Fiasp group (7 events per 100 person-years), both of which were attributed to infusion set failures.
  
  • Dr. Russell offered several notes on these data: (i) the study was not powered to detect a difference in severe hypoglycemia events (Dr. Russell noted that that would have required data from 5,000-6,000 participants over six months); (ii) he believes the greater event rates of severe hypoglycemia in both arms of this study reflect its representativeness; (iii) the total number of events in all groups was still lower than the average in the T1D Exchange registry, which has a more limiting definition of what counts as severe hypoglycemia; and (iv) none of the severe hypoglycemia events were due to device malfunction. Some said that these results led them to wonder whether the bihormonal system might improve upon these hypoglycemia results given that glucagon will be administered, hopefully preventing these severe hypoglycemia events – we’d certainly say we think they will, although even more, we think they’ll

Comparison with other AID pivotal trials

• As noted above, this iLet trial had a different study population, a different methodology (RCT), and different comparator group (any insulin delivery method) compared to other pivotal trials. Although Dr. Russell cautioned against comparing pivotal trials with different methodologies and in a later presentation, Dr. Boris Kovatchev (UVA) implored the audience never to compare CGM outcomes that use different sensors, we’ve collected a brief comparison of some of the key outcomes across AID system pivotal studies – see more on the other AID pivotal trials here. Based on the results summarized below, the Time in Range and A1c improvements seen with iLet appear to be in line with (or above) those seen with AID systems in other trials in adults.  While it may be that the results with children seem slightly more mixed than in other AID pivotals with an A1c improvement similar to that of other trials but about an hour less/day of a Time in Range improvement, we would seek the views of pediatric experts before making such a pronouncement definitively. Overall, we’re thrilled to see that another AID system option may well become available for people with diabetes soon in the US, particularly near-term, given its lower burden setup and bolus use, and longer-term, the opportunity to move hyperglycemia down more easily with the use of glucagon. We do also note we don’t yet have a sense of what degree if much at all the extra “hassle” that may be required with two hormones vs. one – we look forward to learning more from the field on this.

	Adults				Children
	Insulin-only iLet	MiniMed 780G	Control-IQ	Omnipod 5	Insulin-only iLet	Control-IQ	Omnipod 5
Study design basics	~Three-month RCT; compared to G6 + continued baseline insulin delivery (AID, pump or MDI)	Single-arm, compared to SAP or 670G, adults + adolescents	Six-month RCT, compared CIQ to SAP	Three-month single-arm, compared to baseline (18% on MDI, others on pump), adults + adolescents	~Three-month RCT; compared to G6 + continued baseline insulin delivery (AID, pump or MDI)	Ages 6-13; four-month RCT; compared to SAP	Ages 6-14; three -month single-arm, compared to baseline
Endpoint A1c (+/- change)	7.1% (-0.5%)	7% (-0.5%)	7.1% (-0.3%)	6.8% (-0.4%)	7.5% (-0.5%)	7% (-0.4%)	7% (-0.7%)
Endpoint Time in Range (+/- change)	69% (+2.6 hr/day)	75% (+1.4 hr/day)	71% (+2.6 hr/day)	74% (+2.2 hr/day)	60% (+2.4 hr/day)	67% (+3.4 hr/day)	68% (+3.7 hr/day)
Endpoint time <54 mg/dl (+/- change)	Not shared but ~0.3% (+17 sec/day)	0.5% ( -4 min/day)	0.2% (-1 min/day)	0.2% (-1 min/day)	Not shared but ~0.3% (-35 sec/day)	0.2% (-1 min/day)	0.2% (+34 sec/day)

Omnipod 5 associated with +3.4 hours/day Time in Range for type 2s previously on MDI and +5.9 hours/day Time in Range for type 2s previously on basal-only therapy; strong human factors results for first type 2 AID feasibility study

Data from Insulet’s Omnipod 5 AID feasibility study (EP002) in people with type 2 diabetes (n=24) found that the system drove improvements in both A1c and Time in Range after eight weeks of use. Adults with type 2 diabetes previously on MDI (n=12) or basal-only insulin therapy (n=12) with no insulin pump use in the three months before screening were onboarded to Omnipod 5 and used the system for eight weeks at home. As a reminder, the Omnipod 5 system includes (i) a tubeless, wearable insulin pump (the Pod) with embedded algorithm, (ii) a Dexcom G6 CGM, and (iii) Omnipod 5 mobile app that comes pre-downloaded on an Insulet-provided controller or downloadable on a compatible smartphone and the system received FDA clearance for use in people with type 1 diabetes in January and is currently in a limited launch in the US. Participants enrolled in the trial had unrestricted eating and exercise as well as unrestricted use of additional anti-hyperglycemic medications and had an average age of 63 with an average diabetes duration of 19 years. Additionally, 33% of participants identified as Black, 7% identified as Hispanic or Latino, and 4% identified as Asian. Of those enrolled in the MDI arm of the trial 11 were on at least one additional anti-hyperglycemic medication, six were on at least 2 medications, and eight were using either a GLP-1 or SGLT-2. Of those enrolled in the basal-only arm of the trial, 11 participants were on at least one anti-hyperglycemic medication, seven were on at least two, and six were using either a GLP-1 or SGLT-2. Prior diabetes technology use was low among study participants with 54% of patients identified as CGM naïve and 96% of patients identified as pump naïve. Importantly, this is the first study of AID use among patients with type 2 diabetes among the major US pump manufacturers and we are looking forward to seeing data from the six-month study extension presented at ADA 2022.

• Among participants who had previously used MDI therapy, Omnipod 5 was associated with an A1c reduction of 1.2% to 8.1% and a 3.4 hour/day increase in Time in Range to 61% after eight weeks (p<0.05 for both). For those previously on basal-only therapy, Omnipod 5 use was also associated with improvements in A1c and Time in Range with participants seeing an A1c decrease of 1.4% to 8.1% and a Time in Range increase of 5.9 hours/day to 57% (p<0.05 for both). While both groups did fall somewhat short of the 70% Time in Range target, the study had a relatively short duration, and we would not be surprised to see further Time in Range improvements when data from the six-month extension trial is presented at ADA 2022. That said, given that many of the participants were on additional anti-hyperglycemic medications, we wonder what else may be necessary to help patients improve glycemic management and reach Time in Range targets.

• Across both prior MDI and basal-only groups, participants experienced significant reductions in severe hyperglycemia (≥250 mg/dL) with reductions of 2 hours/day to 9.3% and 5 hours/day to 12%, respectively (p<0.05 for both). However, as participants spent little to no time in hypoglycemia, the vast majority of time not spent in range was spent in hyperglycemia (>180 mg/dL) estimated at approximately 39% (7.2 hours/day) and 43% (10 hours/day) at eight weeks for prior MDI and basal users, respectively. Interestingly, prior MDI users saw a significant decrease in total daily insulin dose from 92 units at baseline to 63 units at eight weeks (p<0.05), while prior basal users did not see a significant change in total daily insulin dose. However, given that participants were still experiencing substantial amounts of hyperglycemia while using Omnipod 5, we are curious if this reduced total daily dose was warranted.
• In a separate poster (EP 288), Insulet presented human factors data from 14 adults with type 2 diabetes who participated in the Omnipod 5 feasibility study. Following eight weeks of using Omnipod 5, patients (n=14) participated in 1:1 semi-structured interviews to assess their satisfaction with the Omnipod 5 system and reported a system usability score of 90.5 out of 100. Additionally, patients reported a significant increase in satisfaction with their insulin delivery as assessed with the insulin delivery satisfaction survey, with average satisfaction scores improving from 3.7 at baseline to 4.4 (p<0.05). Overall, participants identified Omnipod 5 as “highly usable,” which is especially notable given that most patients with type 2 diabetes involved in the study were pump naïve.

Insulet Symposium: Dr. Trang Ly shares that type 2s in Omnipod 5 feasibility study performed similarly to A1c-matched participants in type 1 pivotal study; “several hundred” Omnipod 5 users also achieving outcomes mirroring pivotal (+10%-+15% Time in Range) during limited market release; smartphone control with seven Android models, iOS smartphone control prototype now in testing

It was a beautiful Friday evening in Barcelona during the Insulet symposium chaired by Dr. Trang Ly, at which the highlight was undoubtedly the Omnipod 5 type 2 feasibility study results. As a reminder, Insulet’s Omnipod 5 type 2 feasibility study was presented in the ATTD ePoster Hall (EP002) and showed that the AID system was associated with +3.4 hours/day Time in Range for type 2s previously on MDI and +5.9 hours/day Time in Range for type 2s previously on basal-only therapy. In a separate poster (EP 288) at ATTD, Insulet also presented human factors data (n=14) from participants in the Omnipod 5 type 2 feasibility study, showing that participants gave the system a usability score of 90.5 out of 100. A pearl of today’s Insulet symposium was hearing from Dr. Ly that the type 2 feasibility study participants appeared to have performed very similarly to that of A1c-matched participants in the Omnipod 5 pivotal trial, in which people with a baseline A1c of ~8% demonstrated Time in Range improvements to ~60%. The similarities in overnight glucose control between the two cohorts, according to Dr. Ly, are “almost exactly the same,” which she finds “remarkable.” To our knowledge, Omnipod 5’s type 2 feasibility study is the first study of AID use among patients with type 2 diabetes among the major US pump manufacturers, and so we are quite encouraged to hear Dr. Ly share this exciting news that speaks to the potential of AID systems to be beneficial for people with type 2 diabetes on insulin therapy. We are looking forward to seeing data from the six-month study extension phase of the study at ADA 2022, as we imagine that we might see even further improvements in glucose control in this data.

• Dr. Ly noted that the “several hundred” users using Omnipod 5 under limited market release appear to be performing very similarly to participants in the pivotal trial. As a reminder, each Omnipod 5 handheld Controller contains a SIM card that allows for wireless data transmission, and so the company is able to monitor users and assess real-world outcomes. Dr. Ly’s updates are a much-welcomed update following her remarks from Insulet’s recent webinar for advocacy partners, at which she noted that the “several hundred” users of Omnipod 5 thus far have given Insulet a sizable amount of population-level data. Participants in the pivotal trial saw Time in Range improvements from 10%-15%, and so it’s very great to hear that these outcomes are being replicated in the real-world. We continue to think that real-world data are extremely important to validate the strong glycemic management AID can provide and are hopeful that we will be able to see this data in the future.
• Dr. Ly shared notable updates about Omnipod 5’s smartphone control capabilities. As a reminder, Omnipod 5 is the first commercial AID system in the US to offer full smartphone control of basal and bolus insulin, although this is only approved for Android phones thus far. Today, we learned that smartphone control will be compatible with “seven models” of Android phones in the US upon full market availability. This update aligns with CEO Ms. Shacey Petrovic’s comments from our interview with her earlier in the year, during which she said that “there will be a number Android phones that are compatible.” Additionally, Dr. Ly said that she has been testing some prototypes of Omnipod 5 iOS smartphone control, and that the Insulet team has been making “incredible progress.” In the meantime, although iPhone users will not be able to adjust or bolus insulin from their smartphones, the Controller’s SIM connection means they will still be able to see their CGM and pump data from their iPhones in real-time. We know that Insulet has been working to develop iOS smartphone control, and we look forward to seeing this emerge.
• Dr. Anders Carlson (International Diabetes Center), who read out the Omnipod 5 type 2 feasibility data during the symposium, explained how Insulet approached this feasibility study in markedly different ways than its type 1 studies. According to Dr. Carlson, during the Omnipod 5 pivotal trial where ~90% of people were coming from an insulin pump, most participants wanted to understand whether Omnipod 5 would “be the system for them,” and already had a strong baseline understanding of what it means to carb count and bolus coming into the study. As such, Insulet focused on educating that population on the system’s different features. Alternatively, type 2s came in largely tech-naïve, and Insulet instead focused on simplifying the user experience as much as possible not only so that patients find the system easy, but also PCPs, whom, as Dr. Carlson noted, serve the vast majority of people with type 2 diabetes in the US.
• For users wanting to better understand the Omnipod 5 user interface, Dr. Ly shared the Omnipod 5 simulator app (Google Play; Apple Store). After launching the app, users are propelled into a nearly exact user interface that Omnipod 5 users engage with to control and monitor their insulin delivery. Our Associate team enjoyed playing with this app and witnessed dozens of other Insulet symposium attendees immediately download it as soon as it was unveiled.
• This webinar was an absolute success and was highly attended despite ending at 7:45 pm! We greatly encourage you to read our report of Insulet’s exclusive webinar for advocacy partners for more information on Omnipod 5’s limited release, securing coverage for Omnipod 5 from a broad payer network, Omnipod 5’s preschool FDA submission, and integrations with Dexcom G7 and FreeStyle Libre!

MiniMed 780G pivotal trial extension phase with Guardian 4 CGM demonstrates sustained Time in Range and A1c improvements with average Time in Range of 73% and average A1c of 7.1% after three months

Medtronic presented poster data (EP018) demonstrating that participants in the MiniMed 780G pivotal trial who continued to the extension phase and transitioned to the Guardian 4 CGM from Guardian Sensor 3 maintained their glycemic outcomes. As a reminder, Dr. Bruce Bode (Emory University) presented MiniMed 780G’s pivotal trial results at ADA 2020, showing that MiniMed 780G with Guardian Sensor 3 drove +1.4 hours per day Time in Range to 75% among participants (n=157) along with an 0.5% A1c improvement vs. baseline (the results were also published in DT&T). In the ATTD 2022 poster hall, Medtronic showed that children under 18 (n=109) and adults ≥18 (n=67) using MiniMed 780G with Guardian 4 overall maintained an average Time in Range of 73% after three months (children: 72%; adults: 77%). Compared to the overall 75% Time in Range at the end of the MiniMed 780G pivotal trial, this poster shows that strong glycemic outcomes were sustained at three months across the entire study sample, confirming the strong accuracy of the MiniMed 780G AID system with Medtronic’s next generation Guardian 4 CGM. It’s worth noting that the age cohorts of the MiniMed 780G official pivotal were broken down into adults and adolescents, whereas in this group the age stratification for the analysis was done by children and adults, making comparisons between the age-stratified cohorts difficult to do accurately. Nevertheless, we are encouraged by these results as they affirm that Medtronic’s first non-adjunctive CGM is able to support patients using MiniMed 780G to achieve similar glycemic outcomes as were demonstrated in the system’s pivotal trial.

• A1c was sustained from the pivotal trial to the extension phase with Guardian 4 with an average A1c of 7% vs. 7.1%, respectively). After three months of using MiniMed 780G with Guardian 4, children achieved an A1c of 7.2%, whereas adults achieved an A1c of 6.8%.

Abundance of MiniMed 780G real-world evidence: Large cohort (n=25,396) achieves 74% Time in Range and 6.8% GMI; crossover study (n=6,299) boasts +2.7 hours/day Time in Range to 74%; longitudinal cohort (n=9,119) sustains glycemic improvements over six months; Dr. Robert Vigersky highlights potential for clinical inertia around changing AID system settings

Immediately following the Beta Bionics iLet pivotal readout, Dr. Robert Vigersky (Medtronic Diabetes) presented three illuminating real-world studies from MiniMed 780G users. Dr. Vigersky spoke in place of Professor Ohad Cohen, who was originally scheduled to present and who also read out three similarly-designed studies at EASD 2021. We got our first look at real-world data (n=4,120) from MiniMed 780G users one year ago at ATTD 2021 after the system’s OUS launch in October 2020 and CE-Marking in June 2020. In the US, MiniMed 780G is still “under active review” with the FDA after being submitted in February 2021, although this timing is highly subject to the company’s warning letter from the FDA. The studies presented by Dr. Vigersky were compiled from Medtronic CareLink data that was uploaded between August 2020 and January 2022 by users “who provided their consent for data to be aggregated” in countries where local data privacy regulation permits data analysis.

• The entire population of MiniMed 780G users in the study (n=25,396) achieved a 74% Time in Range and 6.8% GMI. These results are extraordinary and add to the already strong data we’ve seen from the system at ATTD 2022, including from two RCTs in the ePoster Hall, a readout from the GIF study, and a readout from the extension phase of the 780G pivotal trial. Notably, 70% of users achieved a GMI <7%, 70% achieved a Time in Range >70%, and 84% saw a Time Below Range (<70 mg/dl) of <4%. Putting these together, 66% of participants achieved both a TIR >70% and a GMI <7%, and 54% managed to secure all three: GMI <7%, Time in Range >70%, and Time Below Range (<70 mg/dl) <4%. Dr. Vigersky also broke down the participants by country of residence, and we’ve included this data in a picture below.

	MiniMed 780G (n=25,396)
Time in Auto Mode	91%
Mean Sensor Glucose	148 mg/dl
GMI	6.8%
Time in Range	74%
Time <70 mg/dl	2.4%
Time <54 mg/dl	0.5%
Time >180 mg/dl	23%
Time >250 mg/dl	5.1%

• Dr. Vigersky then presented a real-world crossover study (n=6,299), demonstrating that users who initiate MiniMed 780G witness a +2.7 hour/day improvement in Time in Range (+11%). Additionally, participants’ GMI fell by 0.3% over the course of the study from 7.2% to 6.9%, and Time Below Range improved by -6 minutes/day from 2.7% to 2.3%. At baseline, 37% of participants achieved a GMI <7% and 33% achieved a Time in Range >70%, and these figures improved to 70% achieving a GMI <7% and 71% achieving a Time in Range >70% after MiniMed 780G initiation.

	Pre-MiniMed 780G (n=6,299)	Post-MiniMed 780G (n=6,299)	Change
Time in Auto Mode	-	91%	-
Mean Sensor Glucose (mg/dL)	163	148	-15
GMI	7.2%	6.9%	-0.3%
Time in Range	63%	74%	+2.7 hrs/day
Time <70 mg/dl	2.7%	2.3%	-6 mins/day
Time <54 mg/dl	0.6%	0.5%	-1 min/day
Time >180 mg/dl	34%	24%	-2.4 hrs/day
Time >250 mg/dl	9%	5%	-1.0 hr/day

• Last, Dr. Vigersky shared data from a longitudinal cohort (n=9,119) showing that the glycemic improvements are observed in the first month after initiating MiniMed 780G and sustained over six months. We are certainly encouraged by these results, as they reinforce the data from the prior two studies and fortify the strong real-world MiniMed 780G outcomes presented at EASD 2021. By the end of the six months, 73% of users attained a GMI <7%, and 73% of users had a Time in Range >70%.

	Month 1 (n=9,119)	Month 2 (n=9,119)	Month 3 (n=9,119)	Month 4 (n=9,119)	Month 5 (n=9,119)	Month 6 (n=9,119)
Time in Auto Mode	95%	94%	94%	94%	93%	93%
Mean Sensor Glucose (mg/dL)	144	144	145	145	146	146
GMI	6.7%	6.8%	6.8%	6.8%	6.8%	6.8%
Time in Range	77%	76%	76%	76%	76%	76%
Time <70 mg/dl	2.6%	2.5%	2.5%	2.5%	2.5%	2.4%
Time <54 mg/dl	0.5%	0.5%	0.5%	0.5%	0.5%	0.5%
Time >180 mg/dl	21%	21%	21%	22%	22%	22%
Time >250 mg/dl	4%	4%	4%	4%	5%	5%

• Importantly, Dr. Vigersky drew attention to clinical inertia around changing AID system settings. In two separate studies of MiniMed 780G users who were not meeting clinical targets for GMI, Time Below Range, and Time Above Range, showing that these users were often using system settings that have been associated with a suboptimal glycemic management. This data further reinforces the sub-analysis of a MiniMed 780G real-world study (n=12,780) that was presented at EASD 2021, showing that even though a lower active insulin time and glucose target correlate with higher Time in Range, only 12% of participants used most aggressive system settings. Of course, we understand that there are many valid reasons for wanting a higher target and a less aggressive active insulin time, especially when it comes to preventing hypoglycemia. Nonetheless, we appreciated Dr. Vigersky’s point that these data highlight a huge opportunity for providers to consider subsets of patients for whom AID system setting adjustments could drive stronger glycemic outcomes. We wonder if Medtronic is thinking about creating a software solution that could identify these patients and provide AID setting adjustment recommendations to HCPs, given the documented association of more aggressive settings with better outcomes.

• In our first look at real-world head-to-head data comparing Guardian 4 and Guardian Sensor 3 in MiniMed 780G, Dr. Vigersky shared that users’ Time in Range, GMI, and Time Below Range were sustained over four weeks of transitioning from GS3 to GS4. This data strongly corroborates Medtronic’s ATTD 2022 ePoster data with extension phase data from the MiniMed 780G pivotal trial, showing that the Guardian 4 CGM demonstrates sustained Time in Range and A1c improvements with an average Time in Range of 73% and an average A1c of 7.1% in the three months after switching from Guardian Sensor 3.

Update on the International Consensus for AID: 40+ KOLs to recommend that AID is “considered” for and made available to for all type 1s; implores all payers to cover or reimburse AID for type 1 diabetes

Rounding out this week of consensus, a Saturday morning session featured an update on yet another consensus report, this time on AID technology. The update was offered by Dr. Moshe Phillip (Schneider Children's Medical Center, Israel), Dr. Revital Nimri (DreaMed), and Dr. Thomas Danne (Auf der Bult Hospital, Germany), who reviewed the consensus meeting that was held at last year’s virtual ATTD sessions and unveiled some of the takeaways of the consensus report that came out of the meeting. The meeting included about 40 KOLs from a dozen countries, a notably greater proportion of whom were women than in the consensus meeting on CGM metrics in clinical trials. The consensus report has seven keys aims: (i) to agree on the indications for use based on clinical studies; (ii) to give recommendations for initiating AID use; (iii) to determine best practices in education, training, and follow-up; (iv) to give clinical guidelines for treatment; (v) to recommend metrics for and presentation of reporting AID data; (vi) to give recommendations related to psychological burden and behavioral challenges; and (vii) to determine the remaining needs and the future of AID systems. To meet these ambitious goals, the panel met virtually in May 2021 and split into nine working groups, each of which focused on one topic within which to discuss existing literature and provide evidence-based recommendations using the ADA’s evidence-based grading system. The nine subgroups covered the introduction; the summary of clinical evidence; the target populations; the initiation of AID use; education, training, and support; clinical recommendations for AID use; data reporting; psychosocial issues and the perspectives of people with diabetes; and the future of AID. The subgroups them presented to the full group for discussion, after which final report recommendations were reached, although these are yet to be published. Below we summarize our key takeaways from this morning’s session.

• The consensus statement will recommend that AID is “considered” for all people with type 1 diabetes. In particular, the group felt that it is hugely important for those facing challenges with general glycemic management, hypoglycemia, and/or significant glycemic variability. They also noted that AID is particularly useful for those with significant hypoglycemia unawareness and frequent or severe episodes. As part of the recommendation that AID should be considered for all people with diabetes, the authors were sure to point out the challenge to overcome racial/ethnic and social inequities driven by unfounded provider biases around who is best suited to AID technology.
• The group felt that there was strong enough evidence to support the use of AID in type 1 adults, adolescents, and school-age children (ages 7-14), with the most efficacy seen in adolescents, those with high A1cs and those on MDI.  However, they felt that there was insufficient evidence on preschool-age children and older adults (ages >65) to currently recommend AID systems’ use in those populations. That said, Dr. Nimri noted that the evidence continues to accumulate, meaning that there very well may come a day where the evidence supports the use of AID in very young and older people with diabetes. Dr. Nimri also highlighted the need for further research into AID use in pregnancy with type 1, in those with type 2 diabetes on basal-bolus therapy, in those with comorbidities (e.g., renal failure), in people with significant hypoglycemia and hyperglycemia, and in people who are not non-Hispanic White. She suggested that further research into AID in these populations could drive an update to the consensus report to include recommendations for use in these populations. Likewise, the consensus group concluded that there are not enough data to conclude that early initiation of AID will preserve beta cell function but do believe that there are “likely benefits” on long-term glycemic control.
• The group also tackled the accessibility front, strongly recommending that AID systems are made available to all people with type 1 diabetes. As a part of this, they recommended that “all the payers (government and private) should reimburse/cover the AID systems along with initial and ongoing AID education and support for the management of T1D.” This is based on evidence that has repeatedly shown that AID provides greater Time in Range improvements than any other current technology. We were pleased to see much discussion around and a direct recommendation of the need for structured training and onboarding, as well as ongoing support, something that Dr. Laurel Messer has reminded us of repeatedly, including at a particularly strong presentation at ATTD 2021.

UK ABCD audit assessing real-world outcomes with DIY AID systems (n=101) finds significant Time in Range (+4.6 hours/day) and A1c (-0.6%) improvements 1.6 years after DIY system initiation

On Saturday, Dr. Thomas Crabtree (University Hospitals of Derby and Burton, UK) shared initial data (n=101) from the Association of British Clinical Diabetologist’s (ABCD) DIYAPS audit program, which aims to capture clinician validated data from DIY AID system users in the real world. At the beginning of his presentation, Dr. Crabtree noted that recent estimates place the number of global DIY AID system users at 2,500, although this is “an absolute bare minimum,” and he would estimate that the true number is far higher. Because these systems are used off label, few studies evaluate their clinical benefit in real-world settings. Because of this, the ABCD DIYAPS audit aims to capture baseline and follow-up data before and after DIY AID initiation on A1c, weight, CGM metrics, hospital admissions and paramedic callouts, and adverse events to better assess the clinical impact of DIY AID systems. The preliminary data presented by Dr. Crabtree include 101 people in the UK on DIY systems with an average follow-up time of 1.6 years. These 101 participants averaged 41-years-old, were 90% White, and had a baseline A1c of 7.0%. Dr. Crabtree said that on average, they already were achieving the consensus A1c goal and that they had long duration of diabetes (26 years). Overall, these DIY AID users saw a significant improvement in Time in Range (+4.6 hours/day; p=0.046) and A1c (0.6 percentage points; p<0.0001), as well as a trend toward reduced time in hypoglycemia and reduced hospital admissions (p=0.063). There were minimal adverse events and no significant change in weight (p=0.9).

	Baseline	Follow-up	Change	p-value
A1c	7.0%	6.4%	-0.6%	<0.0001
Time in Range	56%	75%	+4.6 hours/day	0.046
Time >180 mg/dL	35%	22%	-3.1 hours/day	Not shared
Time <70 mg/dL	9%	3%	-1.4 hours/day	Not significant

New evidence on UVA’s RocketAP algorithm: Hotel-based crossover RCT (n=36) compares Control-IQ vs. RocketAP vs. RocketAP + meal anticipation; RocketAP systems achieve Time in Range of 76% without meal announcements

Dr. Jose Garcia-Tirado (UVA Center for Diabetes Technology) presented the newest data on UVA’s latest AID algorithm, RocketAP. Similar to the most well-known AID algorithm to come out of UVA (Control-IQ), the RocketAP algorithm is a zone-based model predictive controller. However, UVA has added two new features to RocketAP: (i) a new bolus priming system module that is designed to detect unannounced meals quickly and deliver bolus insulin before a lengthy hyperglycemia episode begins; and (ii) a multistage MPC that allows RocketAP to anticipate disturbances in the next two hours. Together, these two features enable RocketAP to better predict and adjust to disturbances (e.g., eating, exercising), thereby reducing disturbances’ impact on glycemic control. Data on the efficacy of the bolus priming algorithm was presented at ATTD 2021 and showed that RocketAP delivered +30% Time in Range over Control-IQ during a six-hour period following an unannounced dinner (n=18). Likewise, initial data (n=18) on the disturbance anticipation multistage MPC was read out at DTM 2021 and showed that the updated RocketAP algorithm that included this feature (deemed “APEX” for the study) conferred an additional +2.4 hour/day Time in Range benefit relative to RocketAP without the disturbance anticipation feature (deemed FCL). During today’s presentation, Dr. Garcia-Tirado read out this trial’s full results (n=36), building on the positive preliminary results read out at DTM 2021.

The randomized crossover study read out by Dr. Garcia-Tirado compared the efficacy of APEX (FCL+) to RocketAP (FCL) and Control-IQ (HCL) in people with type 1 diabetes (ages 18-65) who had used an insulin pump for at least six months and had had diabetes for at least a year. Participants were on average 44 years old,  had had diabetes for 26 years, and had an A1c of 6.7% at baseline (range of 5.1% to 9.1%). Following four weeks of baseline data collection, participants engaged in three 24-hour hospital admissions, during which they were randomized to one of the three systems. During the 24-hour hospital admission, all meals were kept consistent between visits and between participants: dinner was delayed two hours, lunch was fixed, and breakfast was at the expected time based on that individual’s baseline data.

• Although Control-IQ with meal announcements achieved better glycemic control than either RocketAP system, the RocketAP systems still achieved an overall Time in Range of 76% without requiring meal announcements. This is a major finding, as it suggests that with RocketAP and RocketAP + the disturbance anticipation feature, people with diabetes may be able to reap the benefits of AID technology without inputting meal data, bringing AID one step closer to true fully closed loop technology (see our DTM 2021 coverage for more from UVA’s Dr. Breton on fully closed loop systems vs. hybrid closed loop systems). Of course, this was a small (n=36), short (three days) study in a well-controlled environment (hotel) with participants already achieving glycemic targets (baseline A1c averaged 6.7%). Further investigation in a wider population in real life settings would be necessary to truly demonstrate the power of RocketAP.  That said, this early finding is exciting, and we hope it bodes well for further investigations, which are already in the works. It’s also worth noting that the study found no safety concerns related to RocketAP, although this too would need to be confirmed with larger, more representative trials.

• The most statistically significant difference between RocketAP vs. RocketAP + disturbance anticipation (“APEX”) was that participants saw a less overall and nocturnal hypoglycemia with APEX relative to RocketAP. There was also a trend toward a Time in Range improvement following the on-time but unannounced breakfast with APEX relative to RocketAP (63% vs. 58%), although this was not statistically significant. This trend toward higher Time in Range with APEX was accompanied with 50% greater insulin administration in the two hours ahead of the on-time breakfast with APEX as compared to RocketAP and Control-IQ (p=0.02), suggesting that the meal anticipation feature was adjusting its insulin dosing in preparation for the expected meal. While the 63% post-breakfast Time in Range achieved with APEX is still lower than the post-breakfast Time in Range achieved with Control-IQ (75%), this is quite impressive given that participants did not announce this meal with APEX and did announce it with Control-IQ.
• Importantly, when an expected meal did not occur on time (as was the case with dinner in this trial), RocketAP’s disturbance anticipation feature did not result in increased hypoglycemia, as measured by time <70 mg/dL. In fact, compared to RocketAP, the APEX system saw a reduction in time <70 mg/dL post-dinner (2% vs. 0%; p=0.04). This suggests that the meal anticipation feature is safe, even when the anticipated meal does not occur on time.

GIF Study pits Medtronic’s AHCL (780G) and HCL (670G) head-to-head in 7-14-year-olds: Time in Range +1.4 hours/day with 780G and 71% (!) of participants meet consensus Time in Range target

As the headline act of a very busy session hybrid closed loop systems in children, Dr. Revital Nimri (Schneider Children’s Medical Center) read out positive results from the “GIF study” of Medtronic’s Advanced Hybrid Closed Loop (AHCL) and Hybrid Closer Loop (HCL) systems in children ages 7-14 with type 1 diabetes. The study was particularly exciting as AID studies in children are less common than those for adults (though we are seeing more and more) and studies comparing two AID systems head-to-head are rare. The 28-subject study randomly assigned subjects to use AHCL or HCL for six weeks, followed by a 1-2-week washout period, and crossover. During baseline and washout, participants relied on usual care, which was Medtronic’s MiniMed 640G (sensor-augmented pump) for the majority of patients. The two-center study took place in Germany and Israel with fourteen subjects enrolled at each site. As a quick reminder, Medtronic’s AHCL system is commercialized as MiniMed 780G (CE-Marked and launched OUS; submitted to the FDA), and Medtronic’s HCL system is commercialized as MiniMed 670G (launched OUS and US). The AHCL system has a number of improvements over the HCL system, including a lower setpoint (100 mg/dL vs. 120 mg/dL), ability to automatically bolus correction doses, and fewer closed loop exits.

	Baseline	HCL	AHCL
Time in Range	54%	67%	73%
Time <70 mg/dl	2.1%	2.3%	2.8%
Time <54 mg/dl	0.4%	0.5%	0.5%
Time >180 mg/dl	43%	30%	24%
Time >250 mg/dl	14%	9%	6%
Mean glucose	176 mg/dl	159 mg/dl	147 mg/dl
%CV	36.1%	37.6%	37.4%

• Time in Range favored the AHCL system by +1.4 hours/day compared to HCL. On average, users spent 73% of the time in the AHCL phases in range, compared to 67% for the HCL phases. Both systems represented a meaningful improvement over baseline, which was 54% (+4.6 hours/day for AHCL and +3.1 hours/day for HCL). Notably, AHCL outperformed HCL during both daytime and nighttime, though nighttime Time in Range was strong for both groups, as we have seen in past studies of these and other AID systems.

• The majority of the Time in Range improvement was driven by reductions of time in hyperglycemia. Time >180 mg/dl was 43% at baseline, compared to 24% using AHCL (-4.6 hours/day) and 30% using HCL (-3.1 hours/day). By contrast, hypoglycemia metrics were similar across all groups (starting from a low baseline).
• Mean glucose favored the AHCL group by 12 mg/dL (159 vs. 147 mg/dl). Both the AHCL and HCL systems showed a massive improvement compared to baseline, which was 176 mg/dl (-17 mg/dl for HCL and -29 mg/dL for AHCL).  
• During the AHCL phases, 70% of participants met the consensus CGM target for 70% Time in Range. This compares to 29% of participants meeting the same target using HCL, and just 11% of participants at baseline. Put in absolute terms, just three of the 28 subjects had a Time in Range >70% at baseline. When using HCL, this increased to eight participants and when using AHCL, this increased even further to twenty participants.
  
  • This is a very impressive result likely driven by improvements in the AHCL algorithm, as well improved usability with AHCL that increased the time the system was actually in closed loop. During AHCL phases, participants spent a mean of 94% of the time in Auto Mode, compared to 83% with HCL. Similarly, the number of Auto Mode exits was massively reduced from 0.83 per day with HCL to 0.18 with AHCL (~once a day vs. once every five days).

• A1c was reduced from 8.3% at baseline to 7.4% with the use of Medtronic’s AID systems. As each phase of the crossover trial was just six weeks, A1c results were not compared between AHCL and HCL. However, as a population, the participants saw a large reduction in A1c from baseline that was sustained all the way through the three-month extension phase (a total of seven months).

• Total daily dose of insulin was similar for HCL and AHCL phases, but AHCL used a smaller proportion of basal insulin and automatically delivering insulin boluses instead. While 43% of insulin was delivered as basal insulin during the HCL phase, this was reduced to 37% using AHCL. However, the percentage of insulin delivered by user-initiated doses fell from 57% using HCL to 44% using AHCL due to AHCL’s ability to deliver automatic correction boluses.

• The GIF study design and results were very similar to the FLAIR study (read out at ADA 2020), which compared AHCL and HCL in a slightly older group of teenagers and young adults. Overall, both studies demonstrated superiority of the AHCL system on nearly every glycemic and usability metrics - as well as significant improvement from baseline using either system. In general, results from the GIF study appeared to slightly better than those from the FLAIR study, which Dr. Nimri speculated could be due to a higher level of guardian involvement in diabetes management for the younger GIF study population. We enjoyed seeing the results from both studies, which very clearly demonstrate the massive jump that was made with the launch of MiniMed 670G in 2017 and the progress that has been made since then.

Two new RCTs support the direct transition from MDI to MiniMed 780G with hugely impressive glycemic and quality of life benefits: ADAPT study shows six-month 1.4% A1c reduction and 6.6 hour/day Time in Range improvement in those previously on MDI/CGM; separate RCT finds 5.2 hour/day Time in Range improvement and 1 hour/day reduction in time <70 mg/dl on those previously on MDI/BGM

Speaking to a packed plenary hall, Dr. Pratik Choudhary (University of Leicester, UK) and Dr. Thomasz Klupa (Jagiellonian University, Poland) read out the results of two new RCTs on the transition from MDI to MiniMed 780G. In his presentation, Dr. Choudhary read out the highly awaited results of the ADAPT study, a six-month RCT evaluating the safety and efficacy of the direct transition from MDI to MiniMed 780G in adults with type 1 with A1c values >8% using isCGM at baseline. The study included 75 adults (age 40 years) with an average A1c of 9% and average diabetes duration of 18 years. Following Dr. Choudhary's presentation, Dr. Klupa took the stage to present the results of a three-month RCT evaluating a similar transition but in adults with type 1 diabetes on MDI using BGM at baseline. The study included 37 adults (age 40 years) with long-standing diabetes (17 years) and a low average baseline A1c (7.4% in MDI group and 7.1% in MiniMed 780G). Overall, the studies found that there was a significant improvement in Time in Range, A1c, hypoglycemia, and specific quality of life metrics when participants were on MiniMed 780G rather than MDI. Together, these studies support the transition of those on MDI (with or without CGM) directly to MiniMed 780G, disputing the stepwise approach often taken by which people with diabetes are transitioned first to sensor-augmented pump therapy before initiating MiniMed 780G. Based on the studies’ findings, Dr. Choudhary advocated that the use of AID or MDI be a person with diabetes’ choice ­– much like driving, they can choose the management system that works best for them, whether that’s manual or automatic.

• The ADAPT Study showed that those who transitioned from MDI to MiniMed 780G (n=36) saw a six-month 1.4% A1c improvement compared to the group that stayed using MDI and isCGM (n=39), with six-month A1c values of 7.3% and 8.9%, respectively. Notably, this relative improvement was already achieved at three months when those in the MiniMed 780G arm achieved an A1c of 7.3% while the MDI group saw a nonsignificant decline to 8.8% from a baseline of 9%. Furthermore, the MiniMed 780G group saw a whopping 6.6 hour/day increase in Time in Range relative to the MDI group with Time in Range values of 71% and 44%, respectively, at six months, up from 36% and 43%, respectively, at baseline. These massive Time in Range benefits were accompanied by noninferiority in time <70 mg/dl and time <54 mg/dl, which were low at baseline and at six months in both groups (~1-2% and <1%, respectively). Participants in the MiniMed 780G arm spent 96% of time in Auto Mode, and the MDI group used sensors 87% of time. Although most used the optimal settings (glucose target of 100 mg/dl used by 68% and active insulin time (AIT) of 2 hours by 54%), a portion of the participants used less aggressive settings with 32% using a 120 mg/dl target, 41% using an AIT of 2-3 hours, and 5% using an AIT of 3-4 hours. Dr. Choudhary noted that this suggests that the participants saw tremendous glycemic benefits despite many not using the optimal settings.  The total daily insulin dose did not differ between the groups, and there were no major safety concerns.

• • Importantly, a far higher proportion of those in the MiniMed 780G group achieved glycemic goals than did those in the MDI group. Specifically, 53% of MiniMed 780G users achieved a Time in Range >70% compared to only 6.5% in the MDI group. When using the composite goal of Time in Range >70% and time below range <4%, the proportion achieving the goal fell to 39%, although this was still far higher than the proportion in the MDI group (3%).  Likewise, 28% of MiniMed 780G users achieved an A1c <7% (all of whom would have started >8% at baseline) while none of the MDI users did so.

• • At six months, participants in the MiniMed 780G group were significantly more satisfied by their diabetes treatment than the MDI group (p=0.0003 for difference in diabetes treatment satisfaction questionnaire scores). They also saw a far greater improvement from baseline compared to the MDI group (p<0.0001). This is notable because these participants went from using isCGM – which does not require calibration – to a Medtronic sensor that requires calibration, meaning that treatment satisfaction improved despite having to calibrate their CGMs. There was also a decline in hypoglycemia fear un the MiniMed 780G group relative to the MDI group (p=0.04). However, there was no significant difference in overall diabetes quality of life (p=0.2).
  • It’s important to draw attention to the baseline A1c of the study participants, all of which were >8% but the average of which was 9%. This is particularly notable because it reinforces the importance of supporting people with diabetes in using technology regardless of their baseline glycemic management and certainly if what they’re currently using to manage their diabetes is not enabling them to reach targets. That said, an inclusion criteria was scanning ≥5 times a day at baseline (average was ~9 times/day), suggesting that the participants were highly engaged in their diabetes management.
  • It’s also worth noting that the version of MiniMed 780G used during the ADAPT Study is not precisely the same as that which is commercially available. While the algorithm used in the system aligns with that of the commercially available system (with the caveat that the 110 mg/dl target was not available), the hardware and firmware is that of MiniMed 670G. Therefore, it did not have Bluetooth connectivity nor a mobile app.
  • A six-month extension phase followed the study period. Participants in the MiniMed 780G arm continued using the system for six months, and participants in the MDI/isCGM group transitioned to also using MiniMed 780G. The results from the extension phase have yet to be read out.
• The RCT read out by Dr. Klupa found that at three months, adults who transitioned from MDI/BGM to MiniMed 780G (n=20) saw a 0.6% A1c reduction compared to those who stayed on MDI/BGM (n=17). Specifically, those in the MiniMed 780G group saw their A1c fall from 7.1% at baseline to 6.7% at three months, whereas the MDI/BGM group saw their A1c stay 7.4%. The MiniMed 780G group also saw a massive 5.2 hour/day Time in Range improvement relative to the MDI group with Time in Range values of 85% and 62%, respectively, at three months, up from 69% and 63%, respectively, at baseline (p<0.001). The MiniMed 780G group also saw a significant 1.05 hour/day decline in time <70 mg/dl relative to the MDI group (p<0.001). Time <70 mg/dl and time <54 mg/dl was rather high at baseline and in the MDI group at three months, averaging about 5%-6% <70 mg/dl and 2.5%-3% <54 mg/dl. In comparison, time <70 mg/dl and <54 mg/dl dropped to 1.8% and 0.3%, respectively, at three months in the MiniMed 780G group, showing the significant benefit of MiniMed 780G in this population with low average A1c values and long-standing diabetes on MDI and BGM. Importantly, the glycemic improvements seen in the MiniMed 780G group in the first month were maintained through nine months with almost identical CGM outcomes (three additional months of follow-up underway). Participants in the MiniMed 780G group spent an average 98% of time in Auto Mode, and the vast majority used the optimal glucose target (81%) and the optimal active insulin time setting (93%).

• • Perhaps even more striking is the difference in the proportion achieving glycemic targets between the groups. The proportion of participants in the MiniMed 780G group achieving an A1c <7% was far higher than in the MDI/BGM group (75% vs. 29%), despite average baseline A1cs just above 7% in both groups. Perhaps even more impressively, 100% of the participants in the MiniMed 780G group achieved Time in Range >70% while only 29% did in the MDI/BGM group. The difference was even greater when looking at the composite of the Time in Range and A1c targets: 75% of MiniMed 780G users achieved both the Time in Range and A1c targets, whereas only 6% of those in the MDI/BGM group achieved both. This might suggest that those who were achieving an A1c <7% in the MDI group had higher rates of hypoglycemia so that they may not have achieved a Time in Range >70%, whereas those achieving the Time in Range target but not A1c target may have had little hypoglycemia but nearly 30% hyperglycemia. Supporting this potential explanation, 85% of MiniMed 780G users achieved time <70 mg/dl <4%, while only 23% of those on MDI did so.

• Although there was no significant difference in overall quality of life between groups (p=0.29), the MiniMed 780G group did see improvements in the reports of feeling well (p=0.04), of working (p=0.01), of eating as one would like (p=0.01), and of doing “normal” things (p=0.03).

Country-level Time in Range data from MiniMed 780G and MiniMed 770G users demonstrates average Time in Range >70% across geographies; increased time in “SmartGuard” associated with increased Time in Range

In the exhibit hall, Medtronic shared remarkable real-world country-level Time in Range data from MiniMed 770G and MiniMed 780G users around the world. Set up as an interactive display, attendees could choose countries where Medtronic’s AID systems are available and then see average Time in Range across users in the chosen geography. Wow! Population-level data from both MiniMed 770G and 780G was very encouraging, showing that, on average, users are achieving >70% Time in Range on these systems. It was a smart way to set it up, putting the two together …

• Medtronic presented real-world MiniMed 780G population-level data from Italy, the Netherlands, the UK, South Africa, Greece, Poland, and Chile.
  
  • Starting in Italy, Medtronic reported data from 3,584 users who had an average Time in Range of 76%, an average Time Above Range of 12%, and an average Time Below Range of 2%. Wow – all impressive! Patients spent 92% of their time with the SmartGuard feature turned on.
  • In the Netherlands, data from 1,830 users spending 94% of time in SmartGuard were evaluated and demonstrated a Time in Range of 75%, Time Above Range of 22%, and Time Below range of 2.4%. Fancy this, having nearly double TAR (22%!) than Italy (12%!) and roughly the same TBR (2% vs 2.4%).
  • In the UK, data from 996 users who spent 91% of time in SmartGuard demonstrated an average Time in Range of 73%, Time Above Range of 24%, and Time Below Range of 2.3%. Here’s another country with higher TAR – we’d love to see the “over 250 mg/dL”.
  • In South Africa, data from 417 users who spent 89% of time in SmartGuard demonstrated an average Time in Range of 74%, Time Above Range of 24%, and Time Below Range of 2.3%.
  • In Greece, data from 132 users who spent an 95% of time in SmartGuard demonstrated an average Time in Range of 81%, Time Above Range of 15%, and Time Below range of 3.3%.
  • In Poland, 418 users who spent 93% of time in SmartGuard demonstrated an average Time in Range of 81%, Time Above Range of 15%, and Time Below range of 3.3%.
  • Finally, in Chile, data from 165 users who spent an average of 92% of time in SmartGuard demonstrated an average Time in Range of 77%, Time Above Range of 20%, and Time Below Range of 3%.

Across these geographies, there also appears to be an association between increased time in SmartGuard (i.e., when the system is delivering automatic correction boluses) and increased Time in Range highlighting the value of these automatic boluses for helping people stay in range – we aren’t sure this is statistically significant.

Country	Number of Users	Time in SmartGuard (%)	Time in Range	Time Below Range	Time Above Range
Italy	3,584	92%	76%	2.3%	22%
Netherlands	1,830	94%	75%	2.4%	22%
UK	996	91%	73%	2.3%	24%
South Africa	417	89%	74%	2.3%	24%
Greece	132	95%	81%	3.3%	15%
Poland	418	93%	81%	3.3%	15%
Chile	165	92%	77%	3%	20%

• Medtronic presented real-world MiniMed 770G population-level data from the US and Canada. Starting in the US, data from 33,713 users who spent an average of 94% of time in SmartGuard demonstrated an average Time in Range of 71%, Time above Range of 27%, and Time Below Range of 2.9%. In Canada, data from 3,068 users who spent an average of 76% of time in SmartGuard demonstrated an average Time in Range of 70%, Time Above Range of 28%, and Time Below Range of 1.7%. Of note, the SmartGuard in the MiniMed 770G system is the same as that in the MiniMed 670G and does not include automatic correction boluses which is likely related to the slightly lower Time in Range values seen among these patients compared to those on MiniMed 780G.

Country	Number of users	Time in SmartGuard	Time in Range	Time Below Range	Time Above Range
United States	33,713	94%	71%	1.9%	27%
Canada	3,068	76%	70%	1.7%	28%

One-year prospective study from Barbara Davis offers long-term real-world evidence on Control-IQ in youth (n=183): Impressive 1.5% A1c reduction and +3.3 hour/day A1c improvement maintained out to one year in those with baseline A1c values >9%

In an afternoon oral presentation session, Ms. Cari Berget (Barbara Davis) presented real-world data from a prospective, observational study evaluating Control-IQ in youth with type 1 diabetes (n=183) over one year, showing that while those with A1cs >9% maintain their significant benefit over a year, those with lower A1cs don’t always consistently benefit. Ms. Berget noted that there is limited real-world data available on the pediatric use of Control-IQ, although the pivotal trial (read out at ATTD 2020 and published in NEJM in August 2020) showed incredible results with a 3.4 hour/day Time in Range improvement from 53% to 67%. Furthermore, the real-world evidence currently available has been limited to shorter timelines, including three-month pediatric CLIO results and six-month data on the efficacy of Control-IQ combined with a virtual educational camp in children and adolescents with type 1 diabetes (one-year data was read out elsewhere at ATTD 2022), which were both read out at ISPAD 2021. Because of the lack of longer-term data, Barbara Davis found it important to run this study. The study included youth ages 3 to 23 with an average age of 13 whose baseline A1c was 7.6%. For the analysis, youth were stratified into three groups based on their baseline A1c: (i) those with baseline A1c values <7% (n=57); (ii) those with baseline A1c values 7%-8.9% (n=99); and (iii) those with baseline A1c values ≥9% (n=27). Results were provided stratified by baseline A1c; no overall data was read out:

• As might be expected, the greatest benefit of Control-IQ was seen in the group with a baseline A1c >9%. This group saw a 3.4 hour/day Time in Range improvement from 39% at baseline to 53% at three months, which was maintained out to a year. While still not meeting the target for Time in Range, this is a significant benefit in this high-need population. Likewise, this group saw a huge 1.5% A1c improvement from 9.8% at baseline to 8.3% after three months of Control-IQ wear, which was maintained out to one year.
• The group with baseline A1c values of 7%-8.9% offered saw clinical improvements at three months, but these were not entirely maintained out to one year. Specifically, this group saw Time in Range improve from 52% at baseline to 67% after three months of Control-IQ use. However, this fell significantly down to 63% at one year. Likewise, the group saw their A1c fall 0.5% from 7.7% at baseline to 7.2% at three months, but this once again partially reverted at one year, rising a significant 0.3% to 75%.
• Those with A1c values <7% saw a significant improvement in Time in Range from baseline (73%) to three months (79%); however, at one year, their Time in Range fell significantly to 75%, although this was still meeting the consensus target of >70% Time in Range. This group also maintained their already low average A1c throughout one year of Control-IQ wear, increasing nonsignificantly from 6.3% at baseline to 6.4% at three months to 6.6% at one year.

Control-IQ driven Time in Range and A1c improvements maintained among pediatric virtual education camp attendees; Time in Range improvements consistent across children and adolescents

In a Tandem-sponsored session, Dr. Andrea Scaramuzza (Azienda Ospedaliera di Cremona, Italy) presented 12-month data from a prospective, single-arm study evaluating the efficacy of Control-IQ combined with a virtual education camp in children and adolescents with type 1 diabetes. This data is an extension of six-month data we saw presented at ISPAD 2021 and demonstrates maintenance of strong glycemic improvements among the study population. Specifically, among children and adolescents who participated in the virtual education camp and subsequently initiated Control-IQ technology from baseline use of Basal-IQ (n=43) Time in Range improved from 64% to 76% at 1-3 months (p<0.001) and was maintained out to 10-12 months. Similarly, Time Above Range decreased by 6% to 18% at 1-3 months (p<0.001) and stayed at 18% out to 12-months. Time Below Range also improved, decreasing from 9% at baseline to 4% at 1-3 months (p<0.001) and 5% at 10-12 months. Participants also saw sustained improvements in A1c from 7% at baseline to 6.5% at 12 months (p=0.018). **Update: This data was published in JAMA on August 24, 2022 in a research letter entitled "Time With Glucose Level in Target Range Among Children and Adolescents With Type 1 Diabetes After a Software Update to a Closed-Loop Glucose Control System."**

• Excitingly, Time in Range improvements were sustained across age groups including adolescents, who historically have the highest A1c values. Specifically, at 12 months, Time in Range among children 7-11 years old (n=21) was ~76%, and Time in Range for children 12-17 (n=22) was ~74%.

• Dr. Acaramuzza emphasized that Time in Range improvements were sustained across various stages of the COVID-19 pandemic. Specifically, the virtual education camp took place in the summer of 2020 and participants initiated Control-IQ use in the summer and fall of 2020 placing 12-months results in the fall of 2021 in the fourth wave of COVID-19. It is certainly impressive that participants were able to maintain glycemic improvements despite potential COVID-19 related disruptions; however, it is also possible that COVID-19 related restrictions may have limited the activities of participants effectively reducing the number of unpredictable factors that could influence their glycemic levels.

Eight-week RCT finds combining low-dose SGLT-2 (empagliflozin) with Control-IQ improves Time in Range +2.4 hours/day to 81% for adult type 1s (n=32) vs. AID alone

Presenting to a packed room, Dr. Jose Garcia-Tirado (University of Virginia) shared results from the CiQ-SGLT2 study investigating SGLT-2 adjunctive therapy combined with AID in people with type 1 diabetes. The results of this trial were published in DT&T in March 2022. Notably, this is the first free-living trial (no restricted eating or activities) combining a low-dose SGLT-2 with an AID system (Control-IQ) or a predictive low glucose suspend system (Basal-IQ). Adult participants were randomized to receive low dose (5 mg) empagliflozin (n=16) or placebo (n=16) along with sequential Control-IQ (four weeks) and Basal-IQ (two weeks). Participants on empagliflozin and Control-IQ spent 2.4 additional hours per day in Range vs. Control-IQ alone at 81% and 71%, respectively (p=0.04). Likewise, participants on empagliflozin and Basal-IQ spent an additional 4.1 hours per day in Range vs. Basal-IQ alone at 80% and 63%, respectively (p<0.001). Due to time constraints, Dr. Garcia-Tirado focused his presentation on comparing empagliflozin + Control-IQ vs. Control-IQ alone. While the results of empagliflozin plus Basal-IQ are certainly impressive, we understand Dr. Garcia-Tirado’s focus on the Control-IQ arm of the study as the majority of Tandem pump users currently use Control-IQ technology.

• Empagliflozin with Control-IQ led to clinically significant improvements in Time in Range without increasing Time Below Range. Participants on empagliflozin with Control-IQ spent about 16 minutes per day below range (<70 mg/dL) while those on Control-IQ alone spent about 27 minutes per day below range and this difference was not statistically significant. Dr. Garcia-Tirado also shared a CGM trace, highlighting that participants on empagliflozin were more often in range after dinner, compared to those on Control-IQ alone highlighting the ability of SGLT-2s to mitigate postprandial hyperglycemia. 

• While there were no cases of severe hypoglycemia, two participants on empagliflozin discontinued treatment, one due to diabetic ketoacidosis (DKA) hospitalization and one due to dysuria (painful urination). The DKA hospitalization was attributed to a nonfunctioning insulin pump insertion site and SGLT-2 therapy, which is associated with increased DKA risk. The participant who developed dysuria was on empagliflozin with Basal-IQ, and dysuria was resolved through increased fluid intake. Additionally, 13 out of 18 participants on empagliflozin experienced ketosis without DKA, whereas three out of 17 participants on only Control-IQ experienced ketosis without DKA. However, Dr. Garcia-Tirado pointed out that the study was unblinded, so participants who didn’t receive empagliflozin did not measure their ketone levels as often as those who were on empagliflozin. While the prevalence of ketosis without DKA among empagliflozin users is somewhat concerning, we remain optimistic that the advancing field of continuous ketone monitoring may make SGLT-2 therapy for type 1s safer and more accessible.
• Dr. Garcia-Tirado called for developing ketone-aware closed-loop systems that balance ketosis with glucose control. The primary concern associated with the use of SGLT-2s in people with type 1 diabetes is DKA. The three largest trials for SGLT-2 use in patients with type 1 diabetes – DEPICT (for AZ’s SGLT-2 dapagliflozin), inTandem (for Sanofi/Lexicon’s SGLT-1/2 dual inhibitor sotagliflozin), and EASE (for Lilly/BI’s SGLT-2 empagliflozin) – have each resulted in a CRL from the FDA – sotagliflozin in March 2019, dapagliflozin in July 2019, and empagliflozin in March 2020 – largely due to failure to mitigate concerns about DKA. The CiQ-SGLT2 study reduced the dose of empagliflozin to 5 mg to reduce the risk of DKA, but as the publication notes there may be room to further lower the dose and increase Time in Range while further mitigating DKA risk. The publication also notes other strategies to mitigate DKA risk in people with T1D: (i) selecting particular patients with reduced DKA risk; (ii) patient education before initiation of SGLT-2 therapy; and (iii) frequent fingerstick ketone monitoring or continuous ketone monitoring once approved and available.

Real-world Control-IQ data shows massive ~0.8% A1c reduction for MDI users nine months after switching to Control-IQ from 7.9% to 7.1% (n=426); improvements in both pediatrics and adults

During a potpourri of Saturday morning presentation, Dr. Jordan Pinsker (Tandem) presented a first look at nine-month results from the CLIO trial. To start, Dr. Pinsker described CLIO as a “truly real-world use trial,” in contrast to conventional post-market surveillance studies, which usually run through specialized clinical study sites, Tandem recruited participants for its CLIO study by simply sending emails to all customers who purchased a Tandem t:slim X2 pump with Control-IQ. As a reminder, Tandem first began launch for Control-IQ in January 2020. In the email, customers were able to opt-in to the CLIO study and are simply asked to fill out surveys at baseline, 3-, 6-, 9-, and 12-months, while receiving regular care throughout the 12-month study. As a reminder, we’ve seen three-month data from the CLIO study in adults at ATTD 2021, a sub-analysis by racial and ethnic groups at ADA 2021, three-month quality of life data at Keystone 2021, and three-month pediatric data at ISPAD 2021. While this presentation’s results were limited to A1c (and GMI) outcomes, we look forward to Tandem’s future presentations on CGM metrics and other outcomes.

• Results from the cohort of Control-IQ users who switched directly from MDI were particularly impressive with a median A1c reduction from 7.9% to 7.1% after nine months (n=426). This real-world data continues to support the excellent usability of Tandem’s and Dexcom’s devices in the real-world, with no prior pump experience required to see benefits from using Control-IQ. We believe many would be interested in seeing a sub-analysis of users who were CGM-naïve prior to using Control-IQ. For the pump users in the CLIO study, A1c fell from 7.3% at baseline to 7.1% at nine months (n=1,487). Note that baseline figures are lab-measured A1c, while the nine-month outcome is the CGM-derived GMI, meant to provide an estimate for A1c based on CGM-measured mean glucose.

• Excitingly, Control-IQ was shown to be effective at improving A1c in every age group and baseline insulin delivery method. As shown in the table below, in adults, after nine months on Control-IQ, median A1c was universally in the 6.9% to 7.1% range regardless of age and prior insulin delivery method. In particular, we were excited to see this result for the (admittedly small) sample of elderly patients previously on MDI. This is a population where some have the greatest concerns about diabetes technology uptake, but the results from Tandem suggest it can be done successfully. Additionally, Control-IQ delivered significant improvements to A1c in the pediatric group, who generally started and ended with higher A1cs at baseline and nine months.

Diabeloop shares real-world data on its DBLG-1 AID algorithm: +4.4 hours/day TIR to 73% (n=974); ambitious pipeline now includes MDI support and in-house smartwatch with sensors to support fully closed loop

France-based Diabeloop has been a familiar name around the conference circuit over the past few years, but at this year’s ATTD, the company made some of its biggest announcements to-date. CEO Erik Huneker was present at ATTD’s Tech Fair to give an update on the company’s work in and around insulin dosing algorithms. Notably, the company now boasts over 7,000 users on its DBLG-1 algorithm, which works with Dexcom G6 CGM and Roche’s Accu-Chek Insight or Kaleido’s pumps.

• Diabeloop gave a first look at real-world data from 1,914 of the first DBLG-1 users in Germany with mean Time in Range coming in at 73% and over half of users (60%) achieving a Time in Range >70%. GMI for this sample was an impressive 7.1%. In a smaller subset of 974 participants with data from before and after initiation of DBLG-1, Time in Range improved a very impressive +4.4 hours/day following initiation of AID (55% to 73%). Importantly, real-world rates of hypoglycemia using DBLG-1 are also very low. In the full set of 1,917 patients using DBLG-1 in Germany, time below 70 mg/dl was just 0.9%.

• Mr. Huneker also gave an update on Diabeloop’s pipeline, which has become increasingly expansive over the years. In the near-term pipeline, Diabeloop has plans to add enhancements to its algorithm, including self-learning around insulin sensitivity, meals, physical activity, and menstrual rhythms. Diabeloop is also hoping to expand the indication around its DBLG-1 algorithm into children teens and new insulins. Notably, DBLG-1 was the first AID system to receive CE-Marking for use in “highly unstable” diabetes in December 2020. Diabeloop is also planning UI/UX changes by the end of the year, as well as a new machine learning module.
• The two most ambitious items from Diabeloop’s pipeline are its MDI-focused DBL-4Pen platform and its smartwatch for AID systems.
  
  • The DBL-4Pen work was announced by Diabeloop just two weeks ago and aims to integrate Dexcom G6 CGM and Biocorp Mallya’s connected insulin pen attachment with Diabeloop’s smartphone app to titrate insulin doses. Excitingly, Mr. Huneker announced that the first clinical trials for this work are expected to begin in a “few weeks.” Mr. Huneker described his ambition for DBL-4Pen to provide “80% of [the benefits from] closed-loop at 30% of the cost.”
  • In a brand-new “moonshot” item in Diabeloop’s pipeline, Mr. Huneker shared plans to roll-out a “smartwatch” with sensors to help support a fully closed loop AID system. Diabeloop aims to develop this wrist-worn device on its own, and Mr. Huneker shared that Diabeloop has actually been working on this project for the last two years. The smartwatch is expected to be ready sometime “over the next two years.” The custom device is specifically designed to help collect data that will allow Diabeloop to develop and drive a fully closed loop system with no meal or physical activity announcements and even higher Time in Range – Mr. Huneker outlined the targets clearly: 100% time in closed loop with >90% Time in Range. The watch is expected to include sensors for accelerometry and heart rate variability with algorithms embedded in the device to help detect meals and exercise.

• • • The project’s goals, in some ways the “holy grail” for an AID system, are similar to plans shared by Medtronic in the past. In the past, Medtronic has presented its “Personalized Closed Loop” project, which had the ambitious goals for 100% time in Auto Mode and >85% Time in Range. Additionally, in 2019, Medtronic acquired the startup Klue, whose algorithm claimed to be able to identify when and how fast a person is eating and/or drinking using data from an Apple Watch. At the time of the acquisition, Medtronic had aimed to incorporate Klue’s technology into its Personalized Closed Loop system. Of course, we have not heard any updates on Klue or Personalized Closed Loop in quite some time and, particularly with two changes to leadership at Medtronic Diabetes, it’s unclear whether these projects are still in the pipeline.

Omnipod 5 preschool pivotal glycemic target subanalysis: Vast majority of preschoolers used multiple targets; most common targets were 110 mg/dL and 120 mg/dL; lower target associated with higher three-month Time in Range but higher targets saw greater improvement because lower Time in Range at baseline

Dr. Sarah MacLeish (Rainbow Babies and Children’s Hospital) read out subanalysis of the preschool Omnipod 5 pivotal that looked at the glycemic targets most often used by preschoolers in the trial and the relationship between the glycemic target and glycemic outcomes. As a reminder, the preschool pivotal, which was read out at ADA 2021, included 80 children (ages 2-6, average age 4.7 years old) and found that over three months of Omnipod 5 use, Time in Range improved +2.6 hours/day to 68%, up from a baseline of 57%. Omnipod 5 has five customizable targets (110 mg/dL to 150 mg/dL on 10 mg/dL increments) and can be set at different targets at different times of day. Overall, this subanalysis found that participants were taking advantage of opportunity to use different targets for different times of day and that the lower targets were associated with improved Time in Range, as might be expected given they are more aggressive.

• The vast majority (81%) of participants used more than one target throughout the three-month study. When assessed based on the target that participants spent the majority of time (>50% of time), 110 mg/dL and 120 mg/dL were the most-used targets with 35% and 41% of participants, respectively, spending the majority of the trial at these targets. That said, 9% of participants used a 130 mg/dL target a majority of the time and 3% a 140 mg/dl. The remaining 13% had no majority target, as their time was spent relatively evenly split between multiple targets.
• As might be expected, the targets used shifted over time and differed by time of day. Over the three-month trial, the 120 mg/dL and 110 mg/dL targets were most common during the full 24-hour day, although the 130 mg/dL target was as common as the 110 mg/dL at night. The 140 mg/dl and 150 mg/dl were used very little during the day but saw higher usage at night. Over the trial period, there was a major shift in target usage. As shown below, from the first two weeks to the last two weeks of the trial, the use of the 110 mg/dL target increased tremendously, primarily during the daytime. Likewise, the use of the 140 mg/dL target increased over the entire 24-hour day. Use of the 130 mg/dl target declined from the beginning of the trial to the end, potentially because participants became more comfortable with the more aggressive 110 mg/dL target.

• As expected, participants achieved a higher Time in Range when using lower glucose targets; however, the net improvement was greater in those using higher targets due to a lower baseline Time in Range. Time in Range significantly improved from baseline to three months with all targets used, but the highest three-month Time in Range was achieved with the 110 mg/dL and 120 mg/dL targets at 69% and 68%, respectively. That said, because baseline Time in Range was lower with those using the higher targets, the net Time in Range improvement seen was actually greater with higher targets than with lower targets. Using the time-weighted average glucose target, the researchers found that having a time-weighted average glucose target <120 mg/dL was associated a +2.5 hour/day Time in Range improvement from 60% to 71%, while those with time-weighted average targets >120 mg/dl saw a +2.7 hour/day Time in Range improvement from 54% to 65%. This again suggests that use of lower targets enabled a higher endpoint Time in Range to be achieved but that because of a lower Time in Range at baseline, those using higher targets saw a greater net Time in Range improvement. Notably, there was no correlation between time-weighted average target and age or total daily dose (p>0.05 for both). Time in hypoglycemia did not change with the 110 mg/dL but significantly declined with all other targets. Time <70 mg/dl achieved the consensus target of <4% of time at baseline and at three months across the entire cohort and when split by target used.

Real-world patient-reported outcomes from the CLIO study find significant improvements in quality of life, diabetes burden, and device satisfaction after six months on Control-IQ

Dr. Harsimran Singh (Tandem) presented new six-month patient-reported outcomes (PROs) from the Control-IQ Observational (CLIO) study. As a reminder, the CLIO study is a real-world post-marketing study on the efficacy, safety, and quality of life impact of Control-IQ. From the CLIO study, we’ve previously seen three-month glycemic control and quality of life data in adults at ATTD 2021, more in-depth three-month quality of life data at Keystone 2021, a sub-analysis by racial and ethnic groups at ADA 2021, three-month pediatric data at ISPAD 2021, and three-month adverse event rates at DTM 2021.

Dr. Singh shared six-month results on two aspects of psychosocial health: (i) diabetes burden through the Diabetes Impact and Device Satisfaction (DIDS) scale and (ii) diabetes quality of life through the Impact of Diabetes Profile (DIDP) scale. Of the 2,062 participants, 63% were previously on an insulin pump, 36% were previously on MDI, 10% were CGM naïve, and 23% had a baseline A1c ≥8.5%. Notably, the significant benefits of Control-IQ on diabetes impacts, device satisfaction, and quality of life were consistent across previously therapy (pump vs. MDI), baseline A1c (A1c <8.5% vs. A1c ≥8.5%), and age (pediatric vs. adult participants).

• Based on the DIDS scale, participants reported a significant 33% reduction in the impact of diabetes on their overall wellbeing from a score of 4.66 to 3.12 (p<0.001). Dr. Singh highlighted that participants overwhelmingly reported improvement in sleep quality, with a 35% reduction in poor sleep quality due to diabetes. She explained this improvement by pointing out that participants were also significantly less likely to wake up at night to treat low blood glucose. Dr. Singh also expressed enthusiasm about participants’ significantly reduced worry about hypoglycemia and reduced likelihood of missing work or school due to diabetes.
  
  • In terms of device satisfaction, participants reported a 23% increase in device satisfaction from a score of 7.2 to 8.8 (p<0.001). Participants felt significantly more in control of their diabetes and satisfied with their glycemic control. They also reported feeling their device was easier to use than their prior therapy.

• Based on the DIDP scale, participants reported a significant reduction in the burden associated with diabetes management. Dr. Singh said she was most proud of the improvement patients saw in their “freedom to eat as they desired.” She explained that the freedom to eat as desired is the most impacted aspect of qualify of life, irrespective of whether people have type 1 or type 2 diabetes, and this freedom influences many other aspects of quality of life. Participants also reported significantly improved physical health, financial situation, relationships with family/friends/peers, leisure activities, work and studies, and emotional well-being.

Lilly Symposium: Tempo Smart Button CE-Marking and FDA approval anticipated in 2H22; Lyumjev “insulin evolution, not revolution” offers important marginal benefits over traditional rapid acting insulin that translate into clinically meaningful improvements in PROs

Dr. Partha Kar (Portsmouth Hospitals NHS) moderated a session with Dr. Pratik Choudhary (Kings College London) and Dr. Andreas Liebl (Fachklinik, Bad Heibrunn, Germany) reviewing the noted benefits of Lyumjev (ultra-rapid insulin lispro) and the Tempo smart button. Much of the session focused on the marginal benefits offered by these new products that enhance and improve the lives of patients with diabetes through iterative improvements on previously best-in-class products. Dr. Kar made the analogy of newer insulins to newer iPhones, stating that while an older iPhone still gets the job of answering phone calls, the newer versions are faster, more efficient, and smarter, all of which result in a happier customer. Similarly, newer insulins like Lyumjev are faster and smarter while still producing the primary result: better glucose control with a lower risk of hypoglycemia. Data from the PRONTO studies suggest that although there are no major differences in A1c at 26 weeks (see below for a more in-depth overview), patients on Lyumjev saw faster onset of insulin to reduce postprandial glucose excursions, slight reductions in hypoglycemia, +44 minutes/day Time in Range, and increased flexibility in insulin dosing. Ultimately, Dr. Liebl referred to these small improvements as an “insulin evolution, not revolution” and posited that “nearly every person with [MDI-dependent diabetes] would benefit from Lyumjev.”

• Dr. Liebl spent significant time reviewing results from the PRONTO-T1D and PRONTO-TD2 studies, which showed improved outcomes for patients on Lyumjev compared to Humalog. The PRONTO-T2D study found that ultra-rapid-acting Lyumjev was superior to Humalog in controlling one- and two-hour post-prandial glucose excursions during a mixed-meal-test through 26 weeks. Lyumjev was also, again, non-inferior to Humalog on A1c at the end of the study, with the former group reaching an average A1c of 6.92% vs. 6.86% with the latter – both meeting the target A1c goal. While the overall rates of documented and nocturnal hypoglycemia (<54 mg/dL) were nearly identical, Lyumjev conferred significantly higher rates of postprandial hypoglycemia between one to two hours after a meal (0.7 events per patient year vs 0.3, p<0.001) and between two to four hours post-meal (1.0 events per patient year vs. 0.7, p=0.04). The PRONTO-T1D study found that at the end of 26 weeks, mealtime Lyumjev demonstrated a non-inferior A1c reduction vs. Humalog (estimated treatment difference: -0.08%, 95% CI: -0.16-0.00). Moreover, while there was no significant difference in postprandial severe hypoglycemia (<54 mg/dL) between the three groups up to four hours post-meal, mealtime Lyumjev did significantly reduce these events past the four-hour mark (2.72 events per patient year vs. 4.35 with mealtime Humalog (p=0.001) vs. 3.88 with post-meal Lyumjev (p=0.006), which has been attributed to the faster onset and offset of Lyumjev. Based on these positive results, Lyumjev was approved by the FDA in July 2020 for use in adults with type 1 and type 2 diabetes and more recently, was approved for use in pumps in August 2021 based on additional data from the from the PRONTO-Pump-2 trial.
• Importantly, based on data from dQ&A, we can see that these small differences in the data are being translated into improved quality of life for patients living with diabetes. In particular, on metrics that panel participants consistently rank as the most important for mealtime insulin – consistency, amount of glucose control provided, and time to onset – Lyumjev significantly outperforms Humalog, the comparator in the PRONTO trials, as well as Fiasp, the other available ultra-rapid acting insulin on the market. It is worth noting that between 2Q21 and 4Q21, patient satisfaction for Lyumjev increased from 47% to 58%, suggesting an upward trajectory in improved patient outcomes. See the table below for a full breakdown of patient satisfaction, where the percent indicates how many respondents checked “9” or “10” on a 10-point scale (10 is the best).

	Lyumjev (n=30)	Fiasp (n=151)	Humalog U-100 (n=1,332)
Overall satisfaction?	47%	40%	44%
How much glucose control it provides?	67%	40%	36%
How quickly it works?	70%	44%	29%
How consistently it works?	57%	39%	39%
How much hypoglycemia in causes?	50%	28%	26%

• While Lilly did discuss its Tempo Smart Button, the symposium and Q&A offered no updates on system’s launch timing or a clarification on why the system was not CE-Marked by the end of 2021. According to comments from Lilly in May 2021. the company was expecting CE-Mark for the Tempo Smart Button “later in 2021,” although during today’s presentation we saw footnotes on certain slides indicating that the company has indeed not yet received CE-Marking for Tempo. That said, at the ATTD 2022 exhibit hall, we learned that the company is expecting FDA and CE-Mark to both clear by the “end of 2022,” likely at some point in 2H22 (between “summer” and “end of year”), and that the company is planning more trials to evaluate the system that will begin later in the year after the approvals. While we wish we could have learned why Tempo was not CE-Marked earlier, we are sincerely thankful for the exhibit hall representatives and hope that Lilly can navigate the challenges at the FDA and the new MDR rules to secure FDA and CE-Mark approvals by the end of 2022.
• As we learned at EASD 2021, Lilly’s Tempo system, which consists of a Tempo Smart Button, Tempo Pen, and compatible apps, builds on the company’s existing KwikPen infrastructure and stores and transmits data on patient’s insulin dose, timing, and type of insulin delivered. When launched, Lilly’s Tempo smart button will be compatible with Tempo Pens for Abasaglar 100 units/mL, Humalog 100 units/mL, and Lyumjev 100 units/mL. Lilly has also already secured partnerships with Roche, Glooko, myDiabby, and Dexcom to integrate Tempo data across these platforms, giving patients multiple options for how they may choose to view their Tempo pen data.

FDA-stipulated post-approval MiniMed 670G RCT (n=302) validates 670G single-arm pivotal results: Those on 670G see -0.6% A1c improvement and +2.9 hour/day Time in Range improvement relative to those on sensor-augmented pump therapy at six months

Kicking off a valuable oral presentation session focused on AID technology, Dr. Robert Vigersky (Medtronic) read out the results of an RCT evaluating MiniMed 670G vs. sensor-augmented pump (SAP) therapy, which confirmed the clinical benefit of MiniMed 670G demonstrated in the single-arm pivotal. To begin his presentation, Dr. Vigersky noted that the study that he would be reading out is the post-approval RCT that the FDA required when it granted MiniMed 670G its approval in September 2016, an approval which was based on the MiniMed 670G pivotal trial, a single-arm prospective study that compared MiniMed 670G to baseline pump use. Per Dr. Vigersky, the study he read out during this ATTD session is one of three that are being run with MiniMed 670G as part of the post-approval process; however, the other two are currently ongoing.

The six-month MiniMed 670G RCT included 302 AID-naïve participants (ages 2-80, baseline A1c 8.1%) who were randomized to MiniMed 670G (n=151) or SAP therapy (n=151) after two weeks of SAP therapy at baseline. At six months, those on MiniMed 670G saw a 0.6% A1c improvement relative to those on SAP, with those on 670G seeing their A1c improve from 8.3% to 7.3% while those on SAP seeing a slight A1c improvement from 8.1% to 7.7%. In terms of Time in Range, those on MiniMed 670G saw a 2.9 hour/day improvement in Time in Range compared to those on SAP when adjusted for baseline values (p<0.0001). Specifically, those on MiniMed 670G saw their Time in Range improve from 53% at baseline to 67% at six months while those on SAP saw their Time in Range improve only slight from 52% to 55%. Much of this benefit came at night, when those on MiniMed 670G saw their Time in Range improve from 54% at baseline to 74% at six months (52% to 55% among the SAP group), good for a +4.9 hour/day baseline-adjusted Time in Range improvement among 670G users vs. SAP users. While challenging to compare exactly to the 670G pivotal studies, as there were three of those stratified by age group (ages 2-6, 7-13, 14+) while this RCT included participants ages 2-80, these results are generally in line with those of the pivotal trials, providing further confirmation of the value of AID systems, even the first-gen ones.

	Sensor-augmented pump therapy		MiniMed 670G
	Baseline	Six months	Baseline	Six months	Baseline-adjusted mean difference	P-value for between-group difference
Time in Range	52%	55%	53%	67%	+2.9 hour/day	<0.0001
Nocturnal Time in Range	52%	55%	54%	74%	+4.5 hour/day	<0.0001
A1c	8.1%	7.7%	8.3%	7.3%	-0.6%	<0.0001
Glycemic variability (CV)	42%	40%	41%	35%	-5%	<0.0001

• Those with high A1cs saw greater A1c reductions while those with high time <70 mg/dl saw large improvements in time in hypoglycemia. As part of the analysis, the researchers stratified participants into two groups, those with baseline A1c values >8% (n=155) and those with baseline A1c values ≤8% (n=147). Those in the A1c >8% cohort averaged age 35, had had diabetes for 17 years, and had an average A1c of 9.1%, whereas those with A1c values ≤8% averaged age 41, had had diabetes for 23 years, and had an average A1c of 7.2%. As might be expected, those with baseline A1c values >8% saw a far greater A1c improvement than those with baseline A1c values ≤8%. Specifically, those with baseline A1c values >8% using 670G saw their Time in Range improve from 9.2% to 7.7%, good for a 0.8% baseline-adjusted relative A1c improvement in the 670G group compared to the SAP group, which saw its average A1c fall from 9% to 8.2%. Those with baseline A1c values ≤8% still saw a significant improvement relative to the SAP arm (baseline-adjusted relative improvement of 0.3%; p<0.0001). On the flip side, those with baseline A1c values ≤8% saw a greater reduction in time below range than those with baseline A1c values >8%. Specifically, the group with baseline A1c values <8% saw their time <70 mg/dL fall from 8% to 2% (their SAP counterparts saw time <70 mg/dL only fall from 9% to 7%. Those with A1c values ≥8% saw a smaller but still significant time <70 mg/dL improvement from 4% to 2%.
• As was the case in the pivotal trial, MiniMed 670G was found to be very safe with no adverse events occurring among 670G users. There was one DKA event and two severe hypoglycemia events during the run-in period and four severe hypoglycemia events in the SAP arm. 

One-month Time in Range, one-month meal boluses/day, and hypoglycemia fear are 84% predictive of meeting ≥70% Time in Range at one year in children using Control-IQ; new tool based on these results available online for clinical use

In the conference’s first oral presentation session, Dr. Gregory Forlenza (Barbara Davis) presented a model demonstrating that one-month Time in Range, meal boluses/day, and hypoglycemia fear were predictive of meeting a Time in Range ≥70% after 12 months of Control-IQ use. The study, funded by JDRF, aimed to evaluate predictors of 12-month success on Control-IQ in order to better understand what modifiable factors could be addressed by providers to help more people on AID systems achieve the consensus target of ≥70% Time in Range after 12 months of use. Dr. Forlenza and his colleagues based the prediction analysis on data from 162 youth (average age 7.6 years, 7.6% baseline A1c) who initiated Control-IQ use at Barbara Davis. The factors that were considered as potential predictors included baseline characteristics (A1c, type 1 diabetes duration, sex, race, ethnicity, insurance status, prior CGM use, and prior pump use), one-month device use characteristics (sensor wear time, Time in Range, total boluses/day, meal boluses/day, and time in auto mode), and baseline scores on questionnaires (Inspire, PAID, and Hypoglycemia Fear Survey). Among these factors, the only that were predictive of meeting Time in Range ≥70% at one year were: (i) one-month Time in Range; (ii) one-month meal boluses/day; and (iii) hypoglycemia fear based on the Hypoglycemia Fear Survey Helplessness/Worry About Low Blood Glucose score (interestingly, higher hypoglycemia fear was associated with a higher likelihood of meeting the consensus target). Together, these three factors were 84% able to predict meeting the Time in Range goal at one year. Based on the model’s results, the Barbara Davis team created an online clinical tool, into which clinicians and people with diabetes can see how changes in these three factors impacts the probability of one-year success on Control-IQ. Dr. Forlenza was particularly excited to see that two of three factors are highly modifiable (one-month Time in Range and meal boluses/day), providing support the importance of these factors and underscoring the importance of taking action early.

AID in hospitalized patients continues to show improved Time in Range outcomes with no risk of hypoglycemia, even in medically complex patients; future research is needed to fully understand the full clinical impact of inpatient AID

Dr. Charlotte Boughton (University of Cambridge) discussed the in-patient use of AID systems, advocating that they provide the opportunity to transform in-patient diabetes management by improving glycemic management without increasing hypoglycemia and simultaneously reducing the work burden for HCPs. Of concern, hospitalized patients with diabetes have higher rates of infection, longer stays, higher readmission rates, and a 6.4% higher risk of mortality. On top of the daily challenges of managing diabetes, in-patient diabetes management can become even more complicated with changing metabolic responses to insulin, inconsistent oral intake and periods of fasting, and work burden among caretakers. Therefore, Dr. Boughton said there is a significant desire to use new technology, such as AID, to help address some of these challenges. Though the body of literature is still relatively small, recent research has certainly been supportive of AID use in hospitals. Specifically, Dr. Boughton highlighted a 2018 study, which found that using fully closed loop insulin delivery resulted in six additional hours per day in range, lower mean glucose levels and no increased risk of hypoglycemia compared to patients on conventional insulin therapy. Importantly, patients in the study on AID reported greater satisfaction with their glucose levels than patients on conventional insulin delivery, and 98% said they were happy to have their glucose levels automatically controlled by the system.

• Inpatient AID use can be complicated by comorbidities and the complicated and integrated needs of hospitalized patients with diabetes. For example, Dr. Broughton said that irregular eating and fasting patterns can make it difficult to manage glucose levels, especially when patients are receiving nutritional support. However, a 2019 study from Dr. Boughton found that hospitalized patients using AID spent an additional eight hours per day in range compared to those on conventional therapy with no increased risk of hypoglycemia at 68% Time in Range compared to 36% Time in Range (p<0.0001). Another 2019 study found that patients on dialysis saw similar improvements on AID, increasing Time in Range by nine additional hours each day to 69% compared to 32% for those receiving usual care. These results led Dr. Boughton to conclude that AID systems have great potential to improve inpatient management and can be especially beneficial among patients with more complex medical needs.
• Dr. Boughton advocated for additional studies on in-patient AID use to translate what has been demonstrated in early research and literature into routine clinical care. Currently, the CamAPS system is being tested in implementation projects in two NHS trusts to evaluate the efficacy of in-patient AID in a real-world setting. Over the first 90 days of the study, which included 12 in-patients on 10 wards, patients saw an impressive 58% Time in Range with no episodes of severe hypoglycemia or hyperglycemia. Moreover, feedback from the staff and patients has been positive. However, Dr. Broughton reminded attendees that training is crucial for in-patient AID use given that many of the HCPs caring for these patients may not have familiarity with the devices. While the early positive results certainly bode well for the future of in-patient AID in real-world settings, Dr. Boughton said that larger trials and implementation projects will determine whether AID glucose management can actually improve clinical outcomes.

Esteemed Dr. Boris Kovatchev cautions against comparing Time in Range outcomes between AID trials; argues combination SGLT-2/AID therapy as necessary to overcome Time in Range plateau in mid-70% range

Renowned AID expert Dr. Boris Kovatchev (UVA) cautioned that it is unwise to compare Time in Range data across different AID systems. According to Dr. Kovatchev, the “random bias” inherent in each CGM/AID system creates enough variation in sensor readings such that comparing outcomes from different systems and algorithms head-to-head is inaccurate. Dr. Kovatchev illustrated his point by presenting data from a resampling experiment from n=112 individuals (>5.6 million CGM readings), showing the effect that small increments of random slope bias (0%, -2%, -4% … etc.) can have on CGM-derived metrics. Dr. Kovatchev said that two prominent CGMs, Guardian Sensor 3 and Dexcom G6, roughly have a 6% average slope bias, as shown in the FLAIR and iDCL trials. As seen from the table below, an average slope bias of 6% roughly translates to a discrepancy in Time in Range of about 4%, or nearly one hour per day, which is just on the cusp of being considered “clinically significant.” Dr. Kovatchev also said that vast differences in patient follow-up time makes it difficult to do a true “direct comparison” of AID pivotal trial outcomes. As an example, while MiniMed 780G (n=157) with Control-IQ (n=168) had pivotals with relatively similar sample sizes, the spread of the data is quite different (n=610 patient-years vs. n=9,415 patient-years, respectively).

Average CGM Bias	0%	-2%	-4%	-6%	-8%
GMI	7.0	7.0	6.9	6.8	6.7
% Time <54 mg/dL	0.3	0.4	0.4	0.5	0.5
% Time <70 mg/dL	1.6	1.9	2.1	2.4	2.8
Time in Range 70-180 mg/dL	71.4	72.6	74.0	75.3	76.6
% Time >180 mg/dL	27.0	25.6	23.9	22.2	20.6

• Because most commercial AID systems routinely achieve a Time in Range in the mid-70% range, Dr. Kovatchev mused on how to deliver improved outcomes beyond the mid-70s. Dr. Kovatchev noted that Control-IQ, MiniMed 670G, MiniMed 780G, and other “mature” AID systems (e.g., Omnipod 5, CamAPS FX) routinely have users achieving a mid-70% Time in Range and only 1%-2% Time Below Range soon after system initiation, which from our view is quite impressive and a testament to how far AID technology has come since its inception. However, Dr. Kovatchev noted that AID systems currently struggle to deliver further improvements beyond this range. While some AID users see greater improvements from baseline than others, Dr. Kovatchev said that this improvement averages to ~10% Time in Range, and that variations from this mean reflect either stronger or weaker baseline control. As for why this plateau exists, Dr. Kovatchev noted that it is still rather unclear, and that it might be a combination of physiology, behavior, or insulin-control limits. Turning to how people might overcome this Time in Range barrier, Dr. Kovatchev argued that the frontier will be “combination therapy” where people take SGLT-2s while using AID to achieve Time in Ranges above 80%.

Fiasp in MiniMed 780G in children and adolescents (n=30): Preliminary results of the FACT Study don’t offer comparison of Fiasp vs. insulin aspart but do confirm overall benefit of MiniMed 780G in youth

During a Thursday Medtronic-sponsored session, Dr. Klemen Dovc (University Medical Center Ljubljana, Slovenia) shared preliminary results from the Fast Advanced Closed-Loop Therapy (FACT) Study, an investigator-initiated crossover RCT comparing Fiasp vs. standard insulin aspart in children and adolescents with type 1 using MiniMed 780G. Dr. Dovc noted the importance of this study, as the few previous investigations of Fiasp in AID systems were conducted in adult populations – none have evaluated ultra-fast-acting insulins’ potential in AID systems in pediatric populations. This first-in-class study included 30 children ages 10-18 who had had diabetes for ≥6 months, had been on pump therapy for ≥3 months, and had an A1c ≤11%. Participants averaged age 15, a diabetes duration of eight years, and an A1c of 7.5% (range of 5.9%-9.9%). Following one week of sensor-augmented pump therapy with standard insulin aspart, participants began MiniMed 780G and were randomized to either Fiasp or standard insulin aspart. After four weeks, participants switched to the other insulin for another four weeks on MiniMed 780G. In both arms, participants used MiniMed 780G’s optimal settings (glucose target of 100 mg/dL and active insulin time of two hours). There were no dropouts from randomization to study completion.

• The preliminary results that Dr. Dovc presented did not include data comparing glycemic control and safety with Fiasp vs. standard insulin aspart. However, he did present preliminary data comparing baseline to MiniMed 780G with either insulin. Based on the preliminary analysis, MiniMed 780G increased Time in Range by 2.9 hour/day from 66% with sensor-augmented pump therapy at baseline to 78% with MiniMed 780G in the study period. There was also a large difference in time >250 mg/dL, which fell 1.4 hours/day from 9% to 4%. Time below range was not significantly different and on average met the consensus guidelines both at baseline and with MiniMed 780G (~3% time <70 mg/dL). There was no significant increase in total daily dose from sensor-augmented pump at baseline to MiniMed 780G in the study period, and there were no episodes of severe hypoglycemia or DKA. We look forward to seeing the full results and whether Fiasp led to further improvements in these outcomes relative to standard insulin aspart. In particular, we’ll be interested to see how Fiasp impacts post-prandial Time in Range and post-exercise hypoglycemia. Participants were encouraged to be active during the study to assess the value of Fiasp during and after exercise, and based on Dr. Dovc’s commentary, the study captured 500 hours of physical activity data, which will provide ample learnings on the value of Fiasp in exercise.

Sanofi SoloSmart smart button compatible with all Sanofi “Solo” insulin pens; developed in partnership with Biocorp

In a session on connected diabetes pens, Mr. Andreas Bode (Sanofi) unveiled the novel SoloSmart connected button. This novel insulin pen add on is the culmination of Sanofi’s partnership with Biocorp to develop a Sanofi-specific version of Biocorp’s Mallya connected pen cap. Of note, SoloSmart has a new form factor compared to Mallya’s two-part pen cap, but it maintains similar functionalities recording insulin dose, timing, and amount. Additionally, Mr. Bode outlined a connected ecosystem around the SoloSmart connected button including Bluetooth-based data transfer between the cap, a smartphone app, and a Bluetooth enabled BGM. Additionally, Mr. Bode’s connected ecosystem included data upload to the cloud and data sharing with providers to ensure insulin dosing information can be assessed and analyzed to help patients simplify and optimize diabetes management. While Mr. Bode did not give a specific timeline for when SoloSmart would be available for patients, he noted that Sanofi is hoping to bring it to market “soon” and that SoloSmart is compatible with all of Sanofi’s “Solo” insulin pen devices.

Dr. Roman Hovorka shares that an RCT of a fully closed loop CamAPS AID system in type 2 diabetes has completed and been submitted to EASD 2022 for readout in Stockholm

During the closed loop update session, Dr. Roman Hovorka (University of Cambridge) discussed AID technology in type 2 diabetes, which he began by acknowledging that while AID is hugely beneficial in type 1 diabetes, we could better attend to the needs of type 2s and how AID could support them as well. While he didn’t present new data in the talk, he did share a meaningful update: the researchers at the University of Cambridge have completed the analysis of an RCT evaluating a fully closed loop CamAPS system in a broad population of people with type 2 diabetes on basal-bolus therapy. The protocol of this study is included on ClinicalTrials.Gov (NCT04701424) and was previously discussed at ATTD 2021. Excitingly, the study’s results were submitted to EASD for readout at EASD 2022, meaning that we could see the results of the first-ever RCT evaluating AID in type 2s in a mere five months in Stockholm.

Dana Lewis highlights International Consensus Statement on open-source AID systems while emphasizing patient autonomy; argues similarities between DIY AID systems and commercial systems “outweigh differences”; CREATE RCT readout scheduled for ADA 2022 and will mark first large RCT comparing DIY AID vs. SAP

Ms. Dana Lewis, founder of the #OpenAPS movement, discussed the growing body of evidence included in the International Consensus Statement on open-source AID systems, which was published in The Lancet Diabetes & Endocrinology in January 2022. Pushing back on negative attitudes toward open-source insulin delivery automation, Ms. Lewis argued that open-source AID is a safe and effective choice for those who elect to use it as shown by a growing body of scientific evidence. Furthermore, she said that DIY systems have the potential to improve patient outcomes while reducing one’s daily burden of diabetes management. We found Ms. Lewis’s comments to be a natural follow-up from her presentation at DiabetesMine’s Fall Innovation Days 2021, where she noted that “it’s time to modernize our discussion about the risk of diabetes technology” and stop centering discussions on DIY AID systems on the potential for increased risk. As Ms. Lewis sees it, we should weigh the net risk reduction of using DIY AID systems, especially in regard to long term complications, as she argued that DIY AID is far more beneficial than the risk of manually dosing insulin. Turning to the consensus statement, Ms. Lewis explained that the guidelines provide practical guidance on the safe and ethical use of DIY AID from 48 international experts. Importantly, the consensus urges respect for patient (and caregiver) autonomy when it comes to choosing an AID system. The consensus statement encourages providers to familiarize themselves with all AID systems, including DIY systems, and recommends cooperation with, or referral to, other HCPs if they don’t feel comfortable prescribing AID.

• The consensus statement provides specific recommendations for any and all manufacturers of AID systems. Specifically, the statement recommends that: (i) all AID manufacturers fully disclose how their systems operate to enable informed decisions; and (ii) all users should have real-time and open access to their own data.
• On the “quality” of clinical evidence for AID systems, Ms. Lewis highlighted how evidence can come in many forms, including case studies, retrospective studies, in silico simulations, observational studies, comparison studies, and RCTs. Ms. Lewis highlighted that case studies of DIY AID systems have been conducted across a range of populations, including in pregnant people with diabetes and in those who engage in endurance athletics such as marathons and ultra-marathons.
• Excitingly, Ms. Lewis said that results from the CREATE RCT (n=100 type 1s ages 7-70), the first-of-its-kind large RCT studying DIY AID, will be presented at ADA 2022. As a reminder, CREATE (Community deRivEd AutomaTEd insulin delivery) is an open-label, randomized, multicenter, six-month program assessing the glycemic control of OpenAPS compared to sensor-augmented pump therapy. The participants using OpenAPS specifically used the AnyDANA-loop system, consisting of: (i) the OpenAPS algorithm in a locked Android smartphone; (ii) a DANA-i insulin pump; and (iii) a Dexcom G6 CGM. The primary outcome will be Time in Range, and secondary outcomes will look at PROs and platform performance.
• While there are many AID systems with different CGMs, pumps, and algorithms with varying degrees of smartphone compatibility and regulatory approval statuses, Ms. Lewis emphasized that the similarities across these systems outweigh their differences. Ultimately, she underscored the incredible amount of patient choice that these systems offer users and argued that the 40 million+ hours of real-world patient use demonstrate the value of AID, since patients choose to use these systems not just once, but day after day, to better manage their diabetes.
• In closing, Ms. Lewis urged people to modernize how they talk about choosing technology. Echoing her comments from DiabetesMine 2021, Ms. Lewis highlighted how most if not all AID systems, including DIY systems, confer a net risk reduction for people with diabetes compared to manual insulin dosing. During Q&A, we were pleased to hear renowned diabetes advocate Ms. Renza Scibilia (Diabetes Australia) discuss the concept of patient candidacy for diabetes technology, echoing a very insightful talk from Prof. Julia Lawton at DiabetesUK 2022 on the tendency for clinicians to hold prejudicial and erroneous views when evaluating patient candidacy. Ms. Scibilia argued that, rather than asking if a person with diabetes fits into the presumed “model” of ideal candidates for technology, we should reframe the question as: “is the technology suitable for the person with diabetes?”

“It’s a quantum step … this is really, really changing things”: Prof. Julia Lawton shares powerful narratives from the KidsAP02 study of hybrid closed loop in children ages 1-7 (CamAPS FX, Dexcom G6, and Dana pump)

Kicking off an afternoon symposium on use of hybrid closed loop systems in young children, Drs. Julia Ware (University of Cambridge) and Julia Lawton (University of Edinburgh) started on the youngest end of the age spectrum with results from the KidsAP02 study of the CamAPS FX algorithm in children ≤7 years old. Dr. Ware began with a recap of the clinical results from the study, which were first read out at EASD 2021 and later published in The Lancet in January 2022. The topline results are shown in the table below and see our EASD 2021 report for a more detailed breakdown.

Hierarchical endpoints	CamAPS FX (n=73)	Sensor-augmented pump (n=73)	Mean adjusted difference	P-value
Time in Range	72%	63%	+2.1 hours/day	<0.001
Time >180 mg/dl	23%	32%	-2 hours/day	<0.001
A1c	6.6%	7%	-0.4%	<0.001
Mean glucose	146 mg/dl	159 mg/dl	- 13 mg/dl	<0.001
Time <70 mg/dl	4.9%	4.5%	+1 minute/day	0.74

Following Dr. Ware, Dr. Lawton shared qualitative takeaways from interviews from 33 parents of thirty children from the study. These parents represented all four countries where the KidsAP02 study took place and the interviews occurred four months after the conclusion of the trial. During analysis of the interviews, Dr. Lawton’s group identified five themes that emerged: (i) life before using AID; (ii) adjusting to AID; (iii) better control with less work; (iv) facilitating normality; and (v) future optimizations needed.

• Diabetes management prior to the study was nearly universally considered exhausting for parents. Several parents referred to the need for “constant vigilance” over their child’s condition, resulting in poor sleep, employment challenges, and daily anxiety and worry. Additionally, parents shared feelings that their entire lives and conversations were dominated by diabetes and that their kids may be missing out on normal activities, such as sleepovers. Finally, some parents also shared feelings that siblings might be missing out on time and attention due to the overwhelming burden of managing one child’s diabetes.
  
  • “I feel for him… all his friends were going to his friend’s house on Friday and he’s not invited, because obviously, they’re afraid that they don’t know what to do.”
  • “Of course it has a huge impact on your family life, because you’re not focusing on being a family, you’re focusing on …on child to survive, to a certain extent … Then, sometimes, you forget that you have got other kids. And the oldest one has been demonstrating an aggressive behavior, which we believe is caused by the fact that we might have neglected her.”
• During the early period following AID initiation, some parents described a “bumpy ride.” Several of these parents were self-described “micromanagers” and had difficulty adjusting to the AID system and “letting go” of control. Additionally, Dr. Lawton noted that some parents in retrospect believed that they may have compromised the algorithm’s ability to learn and adjust by stepping in and manually dosing insulin.
  
  • “I didn’t feel like I had enough control over it…”
  • “We did sometimes intervene ourselves … in hindsight unfortunately, we should not have done that. The system would have done it … sometimes we meant too well and messed the system up a bit and sent him into a low … this was our mistake.”
• After a transition period, the majority of parents felt that the AID system had delivered better control with less work. This reduced workload was felt by both parents and healthcare providers. On the glycemia front, parents reported more Time in Range and fewer extreme glucose swings. Additionally, parents appreciated the ability for the algorithm to constantly make small insulin dosing adjustments. Parents also reported less need for provider support, a result of confidence in the closed loop system and providers’ abilities to see both glucose and insulin data remotely.
  
  • “The algorithm is giving so many small doses of insulin over a long period of time, that I think like for me that would be so time-expensive to be doing it. I mean, I would never try to emulate exactly what it was doing, because that would just be ridiculous.”
  • “I think that just not having to think about corrections and when to give them, that’s definitely helped… It takes [away] a lot of that thinking of: do I need to tweak his basal? Do I need to change his ratios? Because, it will be working in the background to do all that for you.”
  • “I don’t know what we’re going to talk about this afternoon p[during our review appointment], because there’s nothing to say. Like, everything’s just going really, really well… So, maybe, if we just had it on a basis of, ‘Oh, we’ll definitely have a look at her data every couple of months, and if we feel like there’s a problem, we’ll call at that point,’ and we don’t need a particular scheduled appointment.”
• Using the hybrid closed loop system, parents reported feelings of normality, both for themselves and their families. Parents reported sleeping better and being less worried about their child’s glycemia. Additionally, they reported feeling more confident when their child was in the care of others (e.g., extended family or friends). For their children, the parents reported better mood, concentration, and sleep, as well as the ability to “feel like a child again,” as they were able to attend parties, playdates, and sleepovers.
  
  • “I’d say nighttime is probably the most dramatic difference, because … we probably are only woken by alarms now, like, twice a week. We never set an alarm to test [our child] now and we trust it. Honestly, the line is deadly straight overnight.”
  • “[Before], we had to talk numbers a lot around her. You’re handing over and you’re saying, ‘Oh, she’s got a temp basal at the moment, it’s maybe lasting for two hours to do this’ and she’s hearing this conversation… Now, we just have normal discussions, so it must feel so nice for her… We’ve had more time to just be her parents and to play and do fun things.”
• Finally, some parents did have notes for improvement, particularly around optimizing AID systems for young children. Often, the same parents who struggled relinquishing control to the AID system also felt that the algorithm was too sluggish. Additionally, the CamAPS FX algorithm was hosted on a smartphone and some parents felt that the smartphone was too bulky for small children. Instead, parents suggested a smaller, more age-appropriate device that would still allow for parents to remotely check insulin and glucose data, as well as deliver remote insulin when necessary.
  
  • “Our garden backs onto a cricket pitch and she’ll jump over our wall and go running on the cricket pitch. She doesn’t always stay near her phone and so obviously, when she’s not near her phone, then the Auto Mode bit isn’t working.”
  • “He is a little monkey and he climbs and he does everything … the mobile phone is big and it is rather bulky… so a wristwatch would be, of course, much less bulky and he could move even more freely.

Digital Health, Decision Support, and Telehealth Highlights

Professor Katharine Barnard-Kelly highlights high accuracy of Spotlight Algorithmic Questionnaire (Spotlight AQ) to identify patients’ primary clinical concerns; ongoing RCT with data expected at EASD 2022

Professor Katharine Barnard-Kelly (Barnard Health) discussed extremely positive results from her ongoing Spotlight Algorithmic Questionnaire (Spotlight AQ) study investigating a pre-clinical questionnaire to identify patient priorities and help direct patient-provider interactions. As Prof. Barnard-Kelly explained, provider burn-out is a real and growing problem that impacts the quality-of-care that providers are able to give people with diabetes. Additionally, as patients have limited time with their providers, sometimes as short as 15-20 minutes, Prof. Barnard-Kelly emphasized the importance of identifying and addressing patients’ priority concerns. However, patients can sometimes struggle to articulate these concerns and similarly, providers may not recognize the topics patients want to discuss, leading patients to feeling unheard or misunderstood. To address these challenges and concerns, Prof. Barnard-Kelly and her group developed and are currently studying the use of Spotlight AQ, a pre-clinical questionnaire that patients can take upon arrival in clinic ahead of meeting with their provider. The questionnaire takes <5 minutes to complete, and patient responses are immediately fed into the Spotlight AQ algorithm to identify where patient priorities fall relative to five core domains in clinical diabetes care: (i) glycemic diabetes management; (ii) psychological burden; (iii) social environment and support; (iv) therapy adjustments; and (v) skills gaps and knowledge. These results are then shared with the patient’s provider and can be used to drive more fruitful and personal clinical conversations and care. Spotlight AQ has been in development since 2014 and is a valid tool to assess patient-reported outcomes.

• Spotlight AQ is currently being investigated in a multi-center RCT across seven clinics in the UK. While Prof. Barnard-Kelly was not able to share specific data from this ongoing study, she did share that to date, for the first 50 patients who used Spotlight AQ, the algorithm has accurately identified 100% of patients’ primary concerns. Of note, this initial study population includes patients with type 1 or type 2 diabetes demonstrating the algorithm is effective across both populations. We are extremely impressed by this high accuracy and imagine that Spotlight AQ has helped providers across scenarios better support their patients with personalized and individualized care catered to their concerns and needs that may not always be evidence from glycemic data alone. We also expect that Spotlight AQ may be identifying concerns patients have but have not previously articulated, but may be extremely important for providers to understand such as factors related to caregivers and psychological burden of diabetes. According to Prof. Barnard-Kelly, her group plans to present full results from this ongoing study at EASD 2022 in Stockholm this fall.
• Dr. Barnard-Kelly presented a case study from two patients who presented with similar glycemic data but had greatly differing concerns related to their broader diabetes management. Specifically, while the patients presented similarly in clinical data and thus may have received similar clinical care from their providers, results from Spotlight AQ indicated that the primary concern for one patient was related to social and knowledge components of diabetes management, while the other patient’s primary concern was the psychological burden of diabetes management.

Real-world clinical decision support system use associated with Time in Range improvement of “at least 5%” in 50% of treated patients and improvement of “at least 10%” in 30% of treated patients at six months

Dr. Moshe Phillip (Schneider Children’s Medical Center) presented new real-world data on the use of the endo.digital clinical decision support system demonstrating improved glycemic outcomes among patients. Specifically, endo.digital has been launched in 10 pediatric endocrinology clinics in the US and Israel and was used by providers to care for pediatric patients with diabetes (n=122) on sensor augmented insulin pump therapy with poor glycemic management (defined as Time in Range <45%). Participants were followed for six months, and providers delivered care that was assisted by clinical decision support recommendations. At six months, over 50% of participants saw Time in Range increases of “more than 5%” (1.2 minutes/day) while 30% of participants saw Time in Range increase of “at least 10%” (2.3 hours/day). Additionally, Dr. Phillip shared that 34% of the study population reduced their time >180 mg/dL by “at least 10%” (2.3 hours/day) and that time in hypoglycemia did not change. Of note, Dr. Phillip explained that providers were taking the recommendations of endo.digital as initial recommendations were not repeated by the system in follow-up due to improvements in glycemic management among participants.

• Dr. Phillip highlighted the power of decision support systems to provide specific recommendations such as adjusting basal rates or correction factors at specific times of day. In this real-world analysis, compared to baseline, 75% of participants had their basal rates adjusted in 18 hours of the day while 50% of participants had their basal rates adjusted during all hours of the day. Similarly, 75% of participants changed their correction factors in 14 hours of the day while 50% changed their correction factors in all hours of the day. This level of treatment granularity in terms of the timing of insulin delivery factors is extremely impressive and a strong approach to providing more personalized diabetes care and management.
• Dr. Phillip also discussed behavioral tips and recommendations sharing that endo.digital offers “~130 different behavioral tips” that providers can discuss with their patients. Among the behavioral interventions endo.digital supports, Dr. Phillip highlighted ones related to meal bolus timing, overcorrection of hypoglycemia and subsequent rebound hyperglycemia, and missed meal boluses. At six months, 53% of participants identified as making behavioral changes, though the number and specific types of behavioral changes made was not shared, nor was the impact on TIR or short- or medium-term complications.
• Dr. Phillip emphasized that clinical decision support systems, which have been shown to be non-inferior to expert provider recommendations (see ATTD 2021) do not supersede the decision-making of providers, but can instead act as an additional perspective or option for providers to consider when titrating medications and making changes to treatment regiments. We see this as a very important point as some providers and patients may feel uncomfortable giving over the power of their clinical decision-making to an algorithm. However, as Dr. Phillip explained, decision support systems can “act like another team member” to give providers additional insight and therapeutic recommendations to supplement the treatment and decisions they are already supporting.

DreaMed Advisor Pro users (n=56) in Advice4U RCT and in real world who follow system recommendations are twice as likely to experience clinically significant improvements in Time in Range and Time Above Range

Dr. Revital Nimri (Shneider Children's Medical Center of Israel) shared data showing that type 1s using DreaMed’s AI-based decision support system (n=56) were more likely to see clinically significant Time in Range improvements if they followed system recommendations. As a reminder, DreaMed’s Advisor Pro is a decision support system that offers insulin dosing recommendations based on pump, CGM, and BGM data stored on Glooko or Tidepool’s RPM platforms. We saw the much-awaited readout of DreaMed’s Advice4U trial at ATTD 2020, showing that Advisor Pro use yielded non-inferior results (p<0.0001) to expert physician advice in 122 young type 1s. Then at ATTD 2021, we saw data from Advisor Pro users (n=66) showing that the glycemic benefits and provider satisfaction observed during the Advice4U RCT was also observed in a real-world setting. During this ATTD 2022 session, Dr. Nimri shared retrospective data from young type 1s on pump therapy who received more than one behavioral recommendation from Advisor Pro either in the Advice4U RCT or from EHRs across three US children’s hospitals. The cohort was divided into “followers” (if ≥one behavioral recommendation stopped being presented until the end of the study/follow-up) and “non-followers” (in the case that the same recommendations continued to be provided over the course of the study). In the cohort of 56 participants who received and followed a “bolus timing” recommendation, 32% of followers increased their Time in Range by >5% by the end of the study, and 36% of followers reduced their Time Above Range by >5% at the end of the study. In comparison, among non-followers, only 14% increased their Time in Range >5%, and only 18% reduced their Time Above Range by >5%.

• Dr. Nimri also shared follow-up data from real-world (n=202) and Advice4U (n=41) RCT participants on the proportion of participants who were “followers.” In the Advice4U RCT, 78% of participants who received hyperglycemia recommendations and 56% of participants who received hypoglycemia recommendations were followers. In the real world, only 70% of participants who received hyperglycemia recommendations were followers and 59% of participants who received hypoglycemia recommendations were followers. We’re curious to further understand the factors that make someone more likely (or less likely) to be a follower of decision support system advice. Clinicians continue to emphasize the role of decision support and insulin titration software in reducing provider burden and helping providers who may be less familiar or comfortable with diabetes management offer their patients high quality care. However, we also acknowledge that many individuals might need supplemental support (i.e., coaching from a DCES) to interpret and act on the decision support that they are receiving.

Three LifeScan posters demonstrate improved readings in range, weight loss outcomes, and provider satisfaction from pairing BGMs with digital health technologies

LifeScan presented three posters at ATTD 2022 highlighting the benefit of combining BGM with digital health solutions to help drive improved outcomes. Specifically, LifeScan presented data (EP183) demonstrating an increase in readings in range for patients who used the OneTouch Reveal app along with the OneTouch Reveal meter. Additionally, data from Noom (EP159) demonstrated that among OneTouch Reveal app users, Noom was able to help drive weight loss. Finally, data from providers using the OneTouch Reveal Pro ecosystem during the COVID-19 pandemic (EP203) demonstrated that providers were able to engage with patient data and identify trends to have “more meaningful” conversations with patients.

• After 90-days, use of the OneTouch Reveal mobile app was associated with an increase of readings in range by 8.1% and 11.3% for patients with type 1 (n=4,154) and type 2 (n=13,623) diabetes, respectively. Specifically, LifeScan evaluated readings in range for OneTouch Reveal BGM users who synced their meters to the OneTouch Reveal mobile app over 90-days comparing data from the first 14 days to data from the last 14 days. Users with type 1 diabetes had an average of 58% of their BGM readings in range at baseline and saw this figure improve to 66% of readings in range at the end of 90 days (p<0.005). Users with type 2 diabetes had an average of 72% of their readings in range at baseline and saw this figure improve to 84% at the end of 90 days (p<0.005). Users did not need to log a significant amount of time in the OneTouch Reveal app to see improvements in readings in range. Specifically, among users with type 1 diabetes, two to three app sessions per week (estimated at 11-20 minutes of app-based engagement) was associated with an improvement of readings in range of 7%-8% with “similar trends” observed for people with type 2 diabetes.
• OneTouch Reveal users who also participated in Noom’s weight loss programming and had available weight data saw an average 16-week weight loss ranging from 6.4lbs to 9.9lbs. Four hundred OneTouch Reveal app users joined Noom following the offer of a 16-week free trial building on the partnership between LifeScan and Noom first announced back in May 2020. Of these 400 users, 208 had available weight data and were used in a data analysis to investigate the relationship between app-based interactions with the Noom app and weight loss. Users who had ≥1 Noom app interaction each week for 16 weeks saw the largest average weight loss at 9.9lbs followed by users who messaged with a Noom weight loss coach in eight of the 16 weeks who saw an average weight loss of 9.5lbs. Across all OneTouch Reveal patients in this analysis who joined Noom, interaction with the Noom app drove statistically significantly larger weight loss, indicating that user engagement was correlated with weight loss outcomes.

• Providers who used the OneTouch Reveal Pro ecosystem during the COVID-19 pandemic (n=22) across 21 clinical sites reviewed patient data in the system during visits 45% of the time and ahead of visits 25% of the time. Reflecting on their use of the OneTouch Reveal Pro ecosystem, 55% of providers expressed interest in continuing to use the ecosystem as their standard of care for patients using LifeScan BGMs while 95% of providers agreed that the ecosystem helped them identify patterns and trends leading to more fruitful and productive consultations with patients.

Sharing of device data ahead of clinical visits associated with 1% lower A1c at 8.3% compared to 9.3%; difference driven by provider actions or social determinants of health?

Dr. Mark Clements (Children’s Mercy, Kansas City) presented data demonstrating an association between pre-visit data uploads and reduced A1c and DKA events. As Dr. Clements explained, prior to the COVID-19 pandemic, many patients relied on clinic staff to help upload device data that was then reviewed during visits with providers. However, during the pandemic, the responsibility of uploading this information shifted from clinic staff and toward the person with diabetes and/or their family and caregivers. Evaluating data from patients with type 1 diabetes over 23 years old who had at least one visit with a provider between March 2020 and November 2021 (n=2116) Dr. Clements noted an association between pre-clinic data uploads and lower A1c and rates of DKA. Of note, approximately 50% of patients in this study population shared device data ahead of their clinic visits and were more likely to be younger, white, and non-Hispanic suggesting disparities in device data uploads that could have an impact on diabetes-related management and outcomes. Specifically, patients who shared data ahead of clinic vists had a lower average A1c at 8.3% compared to patients who did not upload data with an average A1c of 9.3%. Similarly, patients who shared device data had a lower rate of DKA episodes reporting only 0.05 episodes compared to 0.16 episodes among those who didn’t share device data. Looking at these data, we are curious if having access to device data allowed providers to better direct their clinical interventions for these patients or if patients who have access to greater resources and are already better able to manage their diabetes are also the ones who are able to upload device data ahead of visits. Dr. Rayhan Lal (Stanford University) raised this question exactly during Q&A musing that if patients have time to upload their data ahead of visits they are also more likely to have greater resources and time available to aid in their diabetes management suggesting that improved outcomes in this population may not be due to a provider’s access to patient data, but rather due to external factors including social determinants of health that can have a significant impact on diabetes management. In response to this question, Dr. Clements shared his view that the difference in outcomes is likely a combination of the two factors and emphasized the importance of enabling passive data sharing to reduce the burden on patients to upload their device data and ensure data is accessible to providers to help them provide the highest quality of care possible.

Differential changes in healthcare visit frequency during COVID-19 pandemic for patients treated in pediatric and adult endocrinology; association between CGM use and increased pandemic visit frequency for adults age 45-64 and older adults (≥65)

Dr. Lori Laffel (Joslin Diabetes Center) presented retrospective data from patients with type 1 diabetes receiving care at the Joslin Diabetes Center (n=~5,500) on the frequency of healthcare visits during the COVID-19 pandemic across age cohorts. Using the large patient base at Joslin, Dr. Laffel presented data from patients across four age cohorts: <18 (n=~800), 18-30 (n=~1,800), 40-64 (n=~2,000), and ≥65 (n=~900) who all received pre-pandemic care at Joslin. Dr. Laffel’s group found that technology and telemedicine use was differentially associated with pandemic visit frequency by age and more specifically, whether patients received care in the pediatric or adult Joslin diabetes clinic. Additionally, among older patients, CGM use was associated with a higher rate of telemedicine adoption shining a light on the value of CGM data for enabling remote patient monitoring and virtual care.

• For patients under 18 years old and using CGM (n=~640), all of whom were treated by pediatric endocrinologists, there was an overall “leftward shift” in the distribution of GMIs toward lower GMI values during the pandemic. Specifically, at baseline in this cohort, 10% of patients had a GMI <7%, 46% had a GMI between 7-<8%, 34% had a GMI between 8-<9%, and 10% had a GMI ≥9%. During the pandemic, this distribution shifted to 18% of patients with a GMI <7%, 44% of patients with a GMI between 7-<8%, 30% of patients with a GMI between 8-<9%, and 8% of patients with a GMI ≥9%. While Dr. Laffel could not confirm correlation, she suggested that this overall decrease in GMI may have been driven by increased visit frequency, enabled by telemedicine, from 2.6 visits/year to 3.2 visits/year during the first year of the COVID-19 pandemic (p<0.001). Notably, visit frequency was the highest among those with baseline A1c values >9% and this subgroup also saw the largest absolute increase in visit frequency from 4.8/year to 5.6 per year (p=0.04). While this finding is not surprising given that patients with higher baseline A1c may be facing additional challenges with their diabetes management necessitating a higher visit frequency, this data does add to the growing body of evidence in support of technology-enabled care for patients not meeting glycemic management targets at baseline.

• For patients between 18-30 years old, Dr. Laffel found a differential impact of the pandemic on visit frequency with those receiving care in a pediatric clinic seeing no change in visits during the pandemic while those receiving care in an adult clinic saw a decrease in visits during the pandemic. Specifically, during the pandemic 93% of visits took place virtually via telemedicine and for patients receiving care in pediatric endocrinology there was no significant difference in visit frequency with 2.8 visits/year pre-pandemic and 2.7 visits/year during the pandemic (p=0.13). However, for people in this cohort receiving diabetes care in an adult endocrinology clinic, visit frequency actually decreased during the pandemic from 2.2 visits/year pre-pandemic to 1.7 visits/year during the pandemic (p<0.001) despite the accessibility of telemedicine enabled care. Interestingly, patients in this cohort also experienced differential changes in visit frequency by A1c during the pandemic depending on where care was received. Prior to the pandemic, Dr. Laffel noted an increase in visit frequency associated with an increase in A1c for all patients aged 18-30 regardless of whether they received care in a pediatric or adult diabetes clinic. However, during the pandemic, patients in this age cohort who received care in a pediatric clinic saw visit frequency increase as A1c increased while visit frequency remained consistent across A1c for those treated in adult clinics (p<0.001) suggesting adult and pediatric care teams at Joslin Diabetes Center may have taken different approaches to telemedicine during the COVID-19 pandemic.
• Among adults and older adults, visit frequency increased during the pandemic with an independent association between CGM use and higher visit frequency. Specifically, for adults age 45-64 visit frequency increased from 2.6 visits/year to 3.0 visits/year and for adults ≥65 visit frequency increased from 2.9 visits/year to 3.2 visits/year (p<0.01 for both). Across these two populations 50% of patients used CGM, which was found to be associated with increased telemedicine use during the pandemic. Dr. Laffel hypothesized that this relationship was likely due to the increased data accessibility and remote patient monitoring capabilities enabled by CGM. Additionally, Dr. Laffel explained that patients who used CGM and participated in telemedicine visits were more likely than their non-CGM using counterparts to continue with telemedicine visits even as in-person visits once again became an option.

Time in Range and Beyond A1c Highlights

One of first therapy industry-sponsored RCTs to use Time in Range as the primary endpoint: InRange RCT finds no significant difference in Toujeo and Tresiba users at twelve weeks (52% vs. 55%), but suggests that further treatment or tech may be necessary to achieve consensus goals

In what was perhaps the headline of the day, a series of KOLs read out the topline results of the Sanofi-sponsored InRange RCT, the first-ever insulin RCT to use Time in Range as a primary endpoint comparing two second-gen basal insulin analogs, which compared Toujeo (insulin glargine U300) and Tresiba (insulin degludec U100) over twelve weeks. The panel included Dr. Richard Bergenstal (International Diabetes Center), Dr. Pratik Choudhary (University of Leicester), Dr. Tadej Battelino (University Medical Center Ljubljana, Slovenia), Dr. Thomas Danne (Auf der Bult Hospital, Germany), Dr. Steven Edelman (UCSD), and Dr. Eric Renard (University of Montpellier, Montpellier, France). The study included 343 people with type 1 diabetes, who averaged 43-years-old and had had diabetes for 20 years. All participants had baseline A1c values of 7%-10% with about two-thirds (62%) of participants having A1c values ≥8%. These 343 participants were randomized to use Tresiba (n=171) or Toujeo (n=172) for twelve weeks, and >95% of participants in both arms completed the treatment and the study. Blinded Dexcom G6 data was collected for 20 days at baseline and for 20 days at 12 weeks. Overall, the study found that there was no significant difference in Time in Range at twelve weeks between Toujeo and Tresiba users, who reached a Time in Range of 52% and 55%, respectively (p=0.055), although we’d note this is still below the consensus target of >70% (more on this below). To Dr. Battelino, these statistically equivalent Time in Range results suggest that the two second-gen basal insulins are “more similar than different.” Likewise, Dr. Choudhary concluded that we can now say that Toujeo and Tresiba are similar in people with type 1 diabetes in terms of Time in Range, glycemic variability, and hypoglycemia. These results build on the Time in Range and glycemic variability outcomes of the observational, retrospective cohort study, OneCARE (n=199), which similarly found that Toujeo and Tresiba led to statistically equivalent Time in Range improvements in real-life practice for type 1s transitioning from first-gen basal insulins to second-gen basal insulins.

Learnings on Time in Range as a clinical endpoint and as a primary outcome

This study has been closely watched ever since experts designed the trial with “Time in Range” as the primary endpoint. Throughout the presentation, the speakers emphasized that this study’s monumental status as one of the first therapy industry-sponsored studies in which Time in Range is a primary outcome, meaning that it offers significant insight for researchers designing future studies in which CGM metrics will be endpoints. The conversation also gravitated to regulators’ views on these outcomes, to which the panelists noted that Sanofi was told not to use Time in Range as its primary endpoint in the study by the FDA and did so anyways, a “bold” move, as Dr. Bergenstal put it. Dr. Battelino drew attention to the consensus meeting on CGM metrics in non-device clinical trials, which was held on Tuesday and during which the FDA participated, and shared his hope that regulators will get on board with the use of CGM metrics in clinical trials. Ultimately, we believe Sanofi will be viewed as a pioneer at this meeting - the results were a level higher in terms of understanding as well as action that can be the next move. Sanofi first brought Lantus to primary care in the early 2000s and now will be able to bring Time in Range data to PCPs in a way that we believe will be quite convincing.

• Dr. Bergenstal noted that this is another step forward toward the full integration and acceptance of Time in Range and CGM metrics. He suggested that three things must happen first: (i) researchers and industry members using CGM metrics in clinical trials; (ii) FDA acceptance and the use of CGM metrics on treatment labels; and (iii) provider and patient belief in the value of CGM metrics. On the latter, Dr. Bergenstal noted that this is already happening, citing data showing that Time in Range is the second most patient-reported factor that has a “big impact” on daily life.
• Later in Friday’s Sanofi-sponsored symposium, Dr. Emma Wilmot argued that the use of Time in Range as a primary endpoint in the InRange study, as well as the real-world study OneCARE and RCT SWITCH PRO, demonstrates the growing recognition of the value of CGM metrics in type 1 diabetes and type 2 diabetes, which will hopefully encourage its further use in clinical trials and as an endpoint in regulatory reviews.

• Given that this study is among the first of its kind, details on the study protocol are certainly important. Beyond the usual protocol elements for a randomized, parallel group study, the study offers an example for running a therapy trial that uses CGM-based outcomes. As noted above, the study used a blinded Dexcom G6 to measure CGM outcomes for 20 days at baseline and at 12 weeks. This is important for two reasons: (i) participants and their providers could not see their CGM data and thereby CGM was not a cofounding intervention; and (ii) 20 days enabled the researchers to ensure they achieved at least 14 days with >70% CGM data, as is the consensus aim. Both groups averaged 16 days of CGM data/participant. A vast majority (80% of participants in the Toujeo arm and for 72% in the Tresiba arm) achieved >14 days of evaluable CGM data (>70% of use). Dr. Bergenstal suggested that this was an important learning point, suggesting that CGM wear time should be greater than 14 days to ensure that enough data is captured. He also suggested that researchers may want even longer CGM wear periods if hypoglycemia is a hugely important outcome. In this study, CGMs were only worn during these two periods, and none of the participants were on personal CGM. We’ll be interested to see if the consensus statement on the use of CGM metrics in non-device clinical trials recommends this approach, particularly given that several of its authors were participating researchers in the InRange study.
  
  • Dr. Choudhary drew attention to several strengths and limitations of the study methodology. Per Dr. Choudhary, the RCT’s methodological strengths include: (i) its prospective, multicenter, active-controlled, parallel-group comparative design; (ii) its evaluation of clinically relevant CGM metrics; (iii) its use of blinded CGM; and (iv) the length of CGM data collection (20 days). He cited two limitations, including the open-label study designs (participants knew which insulin they were receiving) and the lack of longer-term CGM data, which could have provided more robust conclusions. During Q&A, Dr. Edelman noted that the open-label design would have been hard to avoid, particularly because the insulins are at different concentrations (U300 vs. U100), meaning their use would inherently expose which a participant was using.
  • A note on data analysis: While the key takeaway to us was that there was no significant difference in TIR, that analysis was actually the third step in their statistical analysis. First, the researchers determined that Toujeo was noninferior to Tresiba in 12-week Time in Range (p=0.007 for noninferiority), after which they determined that Toujeo was noninferior to Tresiba in glycemic variability at week 12 (p<0.0001 for noninferiority).

Clinical takeaways based in Time in Range and A1c data to inform better care for those on MDI

Beyond the study’s unique trial design, much of the discussion centered on the study’s clinical takeaways. In his section on implications, Dr. Choudhary reminded the audience that despite major advances in diabetes technology like AID, basal insulins are still extremely important today and will continue to be, as the majority of people with type 1 are currently on MDI and many will continue to be due to preference or access. Dr. Choudhary noted that for these people, "We still have a role in optimizing insulins and how we use them,” making these results hugely important. He also highlighted three clinical takeaways, which were expanded upon during the group’s discussion: (i) even with high engagement, close contact with providers, and the newest basal insulins, MDI users are still not achieving the Time in Range target of >70%; (ii) participants were able to achieve a 0.9% A1c reduction with MDI without technology, but that has been harder to achieve in real practice; (iii) those on MDI should potentially administer their basal insulin in the morning rather than the night.

• During the speakers’ presentation and the audience’s Q&A, much discussion centered on the inability of participants to achieve a Time in Range >70% even with close attention, high engagement, and the second-gen basal insulins. All the speakers noted that this fits with data seen elsewhere showing that without technology such targets are hard to achieve, even for those that are highly engaged in their diabetes management. Making a similar argument to the presenters of the FLASH-UK study at Diabetes UK 2022, Dr. Choudhary noted that even CGM might not be enough and that many people with type 1 diabetes may need to use AID to achieve these targets. We’d agree, although we’d also expand the potential solutions to include other treatments, faster prandial insulins, and behavioral changes.
• Dr. Choudhary drew attention to a key finding (discussed below) that it’s possible to achieve a 0.9% A1c reduction in this population using second-gen basal insulins without any technology. He questioned why these results are so hard to replicate in the real world, wondered what about the trial protocol enabled that, and mused on the ways those elements of the protocol could be implemented in the real world.
• During his closing statements, Dr. Choudhary drew attention to the protocol that asked participants to administer their basal insulin in the morning rather than the night as is common practice. Dr. Choudhary suggested that this change may have contributed to the improvement in A1c that was seen in both groups, as morning administration might have been easier to remember and avoids the common nighttime snack taken by people with diabetes who are concerned about overnight hypoglycemia. Based on these results, the speakers mused that perhaps clinicians should be changing their practice and encouraging people with diabetes to take their basal doses in the morning.

More on the study’s outcomes and plans for additional readouts and analyses

• The presenters offered several additional endpoints of interest, although more will be shared in future readouts. At week 12, Toujeo users did not see a significantly higher glycemic variability than the Tresiba users (40% vs. 41%, respectively, p<0.0001). However, neither group achieved the consensus target for a glycemic variability <36%. Based on descriptive analysis, the two groups also achieved similar outcomes for time >180 mg/dl and time <70 mg/dl at 12 weeks: time >180 mg/dl was 42% in the Toujeo group and 38% in the Tresiba group, and time <70 mg/dl was 5.5% among the Toujeo users and 6.5% among the Tresiba users. While the time in hypoglycemia is certainly higher than the target of <4%, Dr. Choudhary noted that this is an expected time <70% for a population not using CGM. In today’s readout, the presenters did not share the baseline CGM metrics for the groups, which would offer a more detailed understanding of the second-gen basal insulins’ glycemic benefits, as well as their mean adjusted treatment difference. However, they intend to share these metrics in future readouts and in the publication. For those seeking A1c data, Dr. Battelino shared that both groups saw a significant reduction in A1c from baseline to twelve weeks. Specifically, Toujeo users saw a 0.8% A1c reduction to 7.4%, and Tresiba users saw a 0.9% A1c reduction to 7.5%. 

• Importantly, there were no significant between-treatment differences in hypoglycemia endpoints, including the incidence and event rate of any hypoglycemia events, severe hypoglycemia events, and CGM-documented hypoglycemia. This was true overall, during the night, and during the day. Furthermore, the safety profiles of Toujeo and Tresiba were consistent with their known profiles, including about one-fifth to one-fourth of patients experiencing a treatment-emergent adverse event (TEAE), 4%-5% reporting a severe TEAE, and <4% reporting a treatment-related TEAE.
• As Dr. Danne noted during Q&A, this data is a “goldmine” for further analysis. While the topline results were read out today, the speakers were clear that much more data will be read out and published. In fact, the group plans to read out post-meal data, AGP profile data (including baseline data), and patient-reported outcomes at ADA 2022 – we cannot wait to see this! Although the timeline was not discussed, the researchers also plan to share BGM data, data comparing hypoglycemia detection with BGM vs. CGM, and outcomes via baseline glycemic variability. Many of these future analyses came up because of audience questions (a full 25 minutes of those) – it was incredible to see the audience so engaged!

Retrospective, cross-sectional study shows Time in Range is inversely associated with prevalence of microvascular complications; mixed findings around macrovascular complications

Dr. Jolien De Meulemeester (KU Leuven, Belgium) shared a retrospective, cross-sectional, real-world study probing the association between Time in Range and long-term complications. In alignment with growing sentiments to look “beyond A1c,” Dr. De Meulemeester kicked off her presentation by touching on the limitations of A1c while heralding the granularity of CGM-derived metrics such as Time in Range, since they allow HCPs to better understand how small-scale patterns contribute to a larger picture of overall glycemia. However, Dr. De Meulemeester said that the association between Time in Range and long-term complications is “still unclear.” As such, she said that the purpose of this study was twofold: (i) to evaluate the prevalence of microvascular and macrovascular complications in relation to Time in Range; and (ii) to define the association between Time in Range and the presence of microvascular and macrovascular complications. This retrospective analysis included type 1s (n=812, mean age 45, 52% male) from the RESCUE and FUTURE studies who had, as Dr. De Meulemeester put it, “suboptimal control” at baseline (A1c 7.9%, Time in Range 53%). Nearly 50% of participants (n=374) developed at least one microvascular complication over the course of the study, with the most common complication being retinopathy (n=280), followed by nephropathy (n=194), and peripheral neuropathy (n=130). Nearly 17% of participants (n=135) developed at least one macrovascular complication.

• Time in Range was inversely associated with the presence of microvascular complications. The investigators broke down the cohort into four Time in Range quartiles: (i) ≤43%; (ii) >43%-≤53%; (iii) >53%-≤63%; and (iv) >63%. Across these four quartiles, those in lower Time in Range quartiles were significantly more likely to have microvascular complications (p<0.001). Broken out by type of complication, a statistically significant trend was most strongly observed between Time in Range quartile and prevalence of retinopathy (p<0.001), but significant associations were also observed for nephropathy (p=0.036) and peripheral neuropathy (p=0.072). For individuals in the lowest Time in Range quartile, the prevalence of at least one microvascular complication was just over 52% (!), whereas in the highest quartile it was 34%.

• There was no association between Time in Range quartile and the overall prevalence of macrovascular complications (p=0.142). However, Time in Range quartiles were associated with the prevalence of “cerebrovascular accidents” (i.e., stroke). We’d note that the length of this study is somewhat fuzzy to us given the retrospective design and that there is still a huge opportunity for more rigorous prospective longitudinal studies that evaluate the relationship between Time in Range and macrovascular complications.

• Dr. De Meulemeester shared two separate models characterizing the association between Time in Range and complications. In model one (blue in the picture below), the investigators corrected for age, gender, diabetes duration, BMI, blood pressure, lipid profile, smoking, lipid lowering agents, and antihypertensive agents. In model two (orange in the picture below), the investigators controlled for all the controls in model one, but also for A1c. As seen in the picture below, after controlling for all the characteristics in model one, Time in Range was significantly associated with composite microvascular/macrovascular complications, all individual microvascular complications, and cerebrovascular accidents. After correcting for A1c, the only significant associations remaining were between Time in Range and retinopathy, and Time in Range and cerebrovascular accidents. Following an audience question from Dr. Marc Breton, an interesting discussion ensued around whether Time in Range encodes additional information beyond A1c in predicting retinopathy and stroke, given the results of model two. Dr. De Meulemeester acknowledged that this is certainly a possibility, and we look forward to continued scholarship in the future that builds on these findings. 

Dr. Irl Hirsch argues A1c, while “critical” for population insights, can be “problematic” and “dangerous” at individual level; lays out vision for future coexistence of A1c and Time in Range; champions CGM for prediabetes diagnosis and management; EDIC post-analysis of adult type 1s (n=765) shows considerable discordance between TIR and A1c

In one of ATTD 2022’s final sessions, Dr. Irl Hirsch (University of Washington) said that while A1c remains “critical” for population-level insights, it can be “problematic” and “dangerous” at an individual level. Despite being scheduled at the same time as several other impactful sessions, Dr. Hirsch’s presentation drew a huge (!) audience. From our view, Dr. Hirsch’s commentary on the “Beyond A1c” movement marks an evolution from his presentation nine months ago at Keystone 2021. At that time, while he still highlighted several limitations of A1c and noted the “crucial” aspects of GMI and Time in Range for clinical decision making, we perceived a new conviction behind Dr. Hirsch’s confidence around CGM-derived metrics as being much more informative measures of glucose control at the individual level.

Today, Dr. Hirsch noted that during the pandemic, HCPs “had no problems” using GMI instead of A1c due to limitations around point-of-care testing, and that most people have no problems using CGM-based metrics for diabetes management. As Dr. Hirsch said: “In our world, we feel the additional granularity makes [A1c] obsolete.” Still, Dr. Hirsch cautioned that it is “unrealistic” to think that every person with diabetes, let alone person with prediabetes, will have access to CGM in every area of the world. The reality is, Dr. Hirsch argued, that “fingerstick glucose monitoring and [A1c] will be around for a long time,” and that “until there is consensus with regulatory agencies that GMI, [Time in Range], and [Time Below Range] should be primary endpoints, it will be difficult to use only CGM for new drug applications.” Of course, we have already seen movement on this front, with an International Consensus group gathering at ATTD 2022 to discuss the standardization of CGM use in clinical trials.

• “So why do we need A1c?” Dr. Hirsch argued that A1c continues to be the only glycemic metric providing a validated assessment of complications and glucose control, and that it is also the most accessible and relatively cheap measure available “for most populations.” As for why GMI should replace A1c, Dr. Hirsch said that current and future CGMs are a more accurate assessment of mean glucose, and that they can provide more detailed information through Time in Range and %CV. 
• Dr. Hirsch laid out a vision for the coexistence of A1c and CGM-derived metrics, as seen in the picture below. The “most important” point of this diagram, according to Dr. Hirsch, is that he firmly believes in prediabetes management and diagnosis through CGM metrics, given the strong body of evidence pointing to GMI-A1c discordance at the individual level. We were fascinated by his opinions. While some would argue that A1c is a sufficient means to diagnose prediabetes, of course we think CGM would be far better, particularly in that over time, using CGM would make it far easier to see who is hurtling toward a diagnosis but don’t have it yet vs. those who are primarily “steady” over time. Ultimately, we believe the field is becoming more and more clear that CGM can offer anyone a wealth of knowledge to understand how their lifestyle affects metabolic health - we were absolutely thrilled to hear Dr. Hirsch’s take on this subject given the enormous importance of pre-diabetes and his persuasive talk on what actions will bring far more focus to this arena. We see steps reducing risk of pre-diabetes as similar to steps being taken currently by many to reduce risks of COVID-19. We hear all the time from people with diabetes that they fear getting long COVID, and so work very hard not to get COVID-19 by wearing masks everywhere and taking every precaution to reduce risk of COVID-19, even though multiple geographies have relaxed recommendations. Similarly, many could begin even more focus on reducing risk of pre-diabetes in order, similarly, to reduce the risk of T2D. While this is not perhaps a perfect metaphor, we note that COVID-19 itself is actually less likely in those without diabetes, so more may be able to achieve two aims at once by working to prevent pre-diabetes (and effectively T2D) as well as COVID-19 (and effectively long COVID).

• • Dr. Hirsch also laid out five requirements to create a “different conclusion” than the one he laid out above over the next five years. According to Dr. Hirsch, CGM needs: (i) better accuracy, especially in the hypoglycemic ranges; (ii) to be more affordable; (iii) better uptake with regulatory agencies; (iv) better uptake with non-diabetologists; and (v) ideally, to be noninvasive. Dr. Hirsch has praised G-WAVE historically – see diaTribe’s “Wave of the Future: New Glucose Technology Could Revolutionize Care.” From our view, we think given arrows, accuracy is pretty strong already; we also think regulatory agencies have been phenomenally supportive of CGM over time, though this has faded in the age of COVID-19, when some of the regulatory resources have certainly gone elsewhere. We’d say that beyond non-diabetologists, many diabetologists themselves could be stronger on uptake – see dQ&A data in people with T2D on insulin for more on this front. To boot, while penetration in T1D has risen at a fast pace in the US and some EU countries, that is not the case in most countries globally, endocrinologist or not. We’d also assert that multiple professional organizations representing PCPs haven’t seemed to have been as enthusiastic as most, even claiming that people with diabetes “can’t understand the complexity” – additionally, we’ve noticed some geriatric experts claiming words to the effect of “all that technology is too complicated, especially for families.” This is obviously absurd, although the cost of course is a major area of unmet need. While volume could take care of some costs moving down, we’d like to see this as a bigger area of focus, and/or we’d like to see manufactures focus more on professional CGM, where CGM could at least be used to “correct” therapy, even if not used more than, say, quarterly, or even once or twice a year to make sure therapy continues to be optimized as much as possible. Regarding non-invasiveness, we’d love to better understand the goals here – if it is that the devices could be less expensive, that’s a solid goal, though if accuracy isn’t as high, we wonder about regulatory enthusiasm, since they may be products that FDA or others could worry about being used by those on insulin where lower accuracy could be problematic. Regarding “ease of use,” we wonder the degree to which current products are effectively seen as nearly non-invasive, since virtually none of the products in this day and age appear to have painful insertions like some did back in the day. Stay tuned – we are excited to have heard this provocative talk though where we land on it is that it’s very critical given how much better many PWD feel, especially those on insulin, who have any sort of continuous system. We remember terming early CGM systems such as Dexcom’s STS, circa 2006, that weren’t that accurate (or reliable, or particularly easy to use relative to today) as quite amazing – they were continuous, which was such a major change! It’s true that was over 15 years ago – we’re staying very tuned to all the improved benefits and features of course and have our minds focused on what can drop prices similar to other new technology like smart phones that have dropped in price. Of course, we also note – six billion smart phones are estimated to be in use today, compared to about six million CGM – we’ve got a long way to go.
• Dr. Hirsch dove into a recently published EDIC post-analysis of adult type 1s with a diabetes duration >35 years (n=765), showing a considerable discordance between Time in Range and A1c. Participants had a mean age of 60, mean diabetes duration of 37 years, mean A1c of 7.8%, and used the FreeStyle Libre Pro for 11.9 days. Dr. Hirsch presented a figure showing a plot of A1c vs. Time in Range matched pairs, and noted that while the trend line reflects the conventional wisdom that a 7% A1c correlates with a 70% Time in Range, there is substantial variation within the sample. Plus, this variation is not unidirectional, meaning that there were plenty of people at a given Time in Range with lower-than-expected and higher-than-expected A1cs than the 7% A1c-70% Time in Range correlation would imply. This paper strongly reflects Dr. Hirsch’s argument that while A1c can be a useful population-based metric, it is incapable of capturing the glycemic profile at an individual level.

• On the validation front, Dr. Hirsch highlighted a table from a recent publication in the Journal of Clinical Endocrinology and Metabolism pointing to the association of Time in Range as a marker of long-term complications. In a sample of n=515 type 1s over two years, individuals with microvascular complications spent one month less Time in Range, on average, compared to those without microvascular complications (p=0.022). Although Time in Range was not associated with a risk of macrovascular complications, it was the only independent risk factor for hospitalizations for hypoglycemia or DKA.

• To illustrate A1c’s inability to serve as a glycemic measure at the individual level, Dr. Hirsch presented a graph of severe hypoglycemia (SH) rates in people over 65 years old. Concerningly, there was a huge rise in SH rates following the ADA and National Committee for Quality Assurance’s (NCQA) guideline updates recommending a treatment goal of A1c <7%. Rates of hypoglycemia only decreased after these guidelines were updated to recommend minimizing hypoglycemia. While attaining an A1c <7% has been rigorously proven to minimize one’s risk of developing diabetes-related complications, Dr. Hirsch noted that this example further reinforces that A1c is most useful when thinking about populations, not individuals. From our view, this figure also reinforces the impact of NCQA guidelines on clinician accountability, and has us thinking of the recently published and Helmsley Charitable Trust-sponsored NCQA white paper, championing CGM-derived metrics and psychosocial outcomes as quality measures of diabetes care.

Time in Range is more strongly correlated with mobility metrics (e.g., risk of falls, aerobic capacity, grip strength) in elderly individuals with type 2 diabetes than A1c

Israel’s Dr. Tali Cukierman-Yaffe (Tel Aviv University) presented results from a very interesting study on the relationship between CGM-measured Time in Range and mobility metrics (e.g., grip strength, risk of falls, etc.) in elderly people with type 2 diabetes. Dr. Cukierman-Yaffe found that while Time in Range was statistically significantly correlated with some mobility metrics, A1c was not correlated with any. In the study, 144 adults, 60 years or older, wore Medtronic’s blinded iPro CGM for one week. Physical assessments were performed at the beginning and the end of the week. These physical assessments included tests validated to measure aerobic capacity, risk of falls, balance, strength, and frailty.

• Time in Range was significantly correlated with grip strength, a measure of risk of falls called timed up and go (involves the participants starting seated in a chair, standing up, walking ten meters, walking back to the chair, and sitting down), six-minute walk test (a measure of aerobic capacity), and a 360-degree turn test (a measure of balance). Higher Time in Range was associated with higher grip strength in the dominant hand and just missed out on statistical significance in the non-dominant hand. Timed up and go was significantly reduced with improved Time in Range, indicating a lowered risk for falls with higher Time in Range values. For every 10% improvement in Time in Range, timed up and go results were 1.65 seconds faster. Additionally, improved Time in Range was associated with improved aerobic capacity (as measured by the six-minute walk test), and improved balance (as measured by the 360-degree turn test).

Test	Beta coefficient	Confidence interval
Grip strength (dominant hand), kg	0.119	0.006, 0.231
Grip strength (non-dominant hand), kg	0.118	-0.003, 0.238
30-second sit-to-stand	0.135	-0.018, 0.288
BERG scale (0-56)	-0.011	-0.174, 0.151
FSST, sec	-0.044	-0.204, 0.116
10-sec walk, m	-0.006	-0.177, 0.166
Timed up and go, sec	-0.165	-0.319, -0.009
Six-minute walk, m	0.169	0.023, 0.313
One leg stance, sec	0.105	-0.064, 0.273
360-degree turn test, sec	-0.164	-0.319, -0.008

• Dr. Cukierman-Yaffe’s study adds on to the evidence base validating Time in Range as an important metric (see a set of posters from Day #1 for more). Of course, it is still unclear whether higher Time in Range is a symptom or a driver of improved mobility metrics in this population. Regardless, the results demonstrate that we are likely just beginning to scratch the surface of information with CGM. As CGM becomes more widely used across broader populations, the role of glycemia in people’s everyday lives, for those with and without diabetes, may become more evident. To that end, we recall a study presented by Dexcom at ADA 2021, which demonstrated a link between CGM-measured mean glucose and the prior night’s sleep in people without diabetes.

Temporal CGM analysis over course of pregnancy shows that maternal glucose levels between weeks 20-30 may play a large role in large for gestational age infant risk; LGA infants exposed to slightly higher glucose levels throughout entire pregnancy from week 10 on

During a session dedicated to diabetes technology and pregnancy, Prof. Eleanor Scott (University of Leeds, UK) discussed the results of an analysis comparing CGM metrics over the course of pregnancy for pregnant people who had large for gestational age (LGA) babies vs. those whose babies were not LGA. Prof. Scott’s presentation offered our first look at this study, which included data from 386 pregnant people (200 from CONCEPTT and 186 from a Swedish observational study) and will be published in Diabetes Care. Prof. Scott and colleagues were driven to conduct the analysis because summary statistics like full-pregnancy Time in Range and A1c sometimes miss temporal differences that matter when it comes to pregnancy outcomes. For example, although Time in Range and A1c improved in the CGM group relative to the non-CGM group in the CONCEPTT study, more than 50% of the CGM group still had LGA babies. Prof. Scott argued that looking at participants’ CGM metrics continuously throughout the course of a pregnancy can better explain these patterns and provide more useful insights into key preventative actions.

• Across the three metrics evaluated (average glucose value, time in pregnancy range (70 mg/dL-140 mg/dL), and time above range >140 mg/dL), the maternal glucose trajectory was quite similar between those with LGA and with non-LGA babies until week 10. However, after week 10, the maternal glucose trajectories of those who would have LGA babies diverged from those with normal weight babies. Interestingly, across the three metrics (mean glucose, time in pregnancy target range, time above pregnancy target range), those who have LGA babies see their glycemic improvements revert between weeks 20-30 while those with non-LGA babies saw continued decline. Per Prof. Scott, this suggests that glycemic management between weeks 20-30 may has a pathophysiological role in determining whether a pregnant person has a LGA baby.  
  
  • Looking at mean glucose, glucose levels fall quickly until week 10 irrespective of baseline maternal glycemia or infant size, but then diverge for those with LGA vs. non-LGA babies. Overall, LGA infants were exposed to slightly higher levels of glucose for sustained, long periods of time throughout the pregnancy and a much higher glucose level between week 20-30.
  • For both those with LGA infants and with non-LGA infants, time in pregnancy target range increases across pregnancy with a plateau between 10 and 25 weeks for those with non-LGA infants and a decline in Time in Range among those with LGA babies until week 30. Notably, the majority of these pregnant people on CGM and intensive insulin therapy only reached the consensus target for time in pregnancy target range ≥70% at week 34, suggesting much more may need to be done to make that target feasible.
• Because these results indicate what glycemic levels put a pregnant person at higher risk for a LGA baby, they provide insights into the glycemic goals that should be set at different times in pregnancy. For example, based on this data, Prof. Scott suggested that it’s okay to have a higher mean glucose and a lower Time in Range in those first 10 weeks (both those with and without LGA were not meeting targets at that point), but that beginning around week 10, pregnant people with diabetes should aim for a mean glucose value ≤126 mg/dL and a time 70 mg/dL-140 mg/dL of 55%-60% to lower the risk of LGA. Based on these results, Prof. Scott advocated for weekly CGM-metric targets for each week of pregnancy to more effectively address the risk for LGA.

“To Infinity and Beyond” A1c: Posters Challenge Previous Knowledge and Further Propel Discussion Around CGM-Based Metrics and Patient-Reported Outcomes

This modified catchphrase of Space Ranger Buzz Lightyear echoed in our heads as we browsed ATTD’s expansive poster hall on day 1: “To infinity and beyond A1c.” Following a busy and insightful meeting on CGM metrics in clinical trials just yesterday, we felt inspired as we viewed several posters related CGM-based metrics and the Beyond A1c movement. We’ve highlighted some of the most notable ones below and let us know if we missed your favorite.

• First off, work from the University of Colorado’s Barbara Davis Center demonstrated the relationship between Time in Range and A1c (EP105). Of course, the commonly accepted rule of thumb is that a 10% increase in Time in Range is linked to a 0.5% reduction in A1c – this was published by Dr. Roy Beck et al. in 2019. However, this rule of thumb was derived from four clinical trials; the poster authors put that rule of thumb to the test using real-world data from 542 CGM-using adults with type 1 diabetes. Results from the real-world were slightly different, as each 10% increase in Time in Range was associated with a 0.34% reduction in A1c. However, the real-world data also affirmed another rule of thumb that a 70% Time in Range corresponds to an A1c of 7%. In the University of Colorado dataset, participants with 70% Time in Range had a mean A1c of exactly 7.0%.

Time in Range	Real-world data A1c	Beck et al., 2019 A1c
20%	8.7%	9.4%
30%	8.3%	8.9%
40%	8.0%	8.4%
50%	7.7%	7.9%
60%	7.3%	7.4%
70%	7.0%	7.0%
80%	6.7%	6.5%
90%	6.3%	6.0%

• A similar second poster from the University of Colorado looked for the appropriate glucose variability target (%CV) to minimize hypoglycemic risk (link to poster). Of course, the consensus target for glycemic variability is a %CV of 36% or less, as defined by the 2019 CGM consensus targets. Using the same set of 542 CGM-using adults with type 1 diabetes, the poster authors found that the consensus target of %CV of 36% or less may not be aggressive enough to avoid hypoglycemic events. Compared to those with a %CV >40%, the relative risk reduction for level one hypoglycemia (<70 mg/dL) was just 11% for those meeting the consensus <36% target (not statistically significant). This relative risk reduction associated with level two hypoglycemia (<54 mg/dL) was considerably better at 62%. As CGM utilization continues to grow, we look forward to seeing increasing data around CGM metrics and continued refinement of the consensus targets. While this study was focused on adults with type 1 diabetes, we are also looking forward to building the clinical evidence base around other populations such as pregnant people, pediatrics, and people with type 2 diabetes.

	Relative risk
Glucose variability (%CV)	Level 1 hypoglycemia (<70 mg/dL)	Level 2 hypoglycemia (<54 mg/dL)
<30%	0.57	0.14
30%-33%	0.74	0.28
>33%-36%	0.83	0.32
>36%-40%	0.89	0.50
>40%	1	1

• On the Time in Range validation front, a poster out of the University of Antwerp used data from the RESCUE, FUTURE, and ALERTT1 studies to demonstrate a correlation between increasing Time in Range and decreased risk for retinopathy (EP144). The analysis broke out 328 adults with type 1 diabetes into three groups: a group that saw mean Time in Range decrease by >10% over two years (n=65), a group that saw mean Time in Range change by less than 10% over two years (n=201), and a group that saw mean Time in Range increase by >10% over two years (n=62). In the group with reduced Time in Range, the rate of retinopathy was 61%, compared to 51% and 42%, respectively, for the other two groups. Additionally, these changes in Time in Range were significantly correlated with systolic blood pressure. The authors also found that Time in Range was not significantly correlated with rates of neuropathy or microalbuminuria; however, CGM-measured mean glucose was significantly associated with these two complications.
• Finally, on the patient-reported outcomes side, a poster from the University of Leeds (LB044) found an association between improved Time in Range and sleep quality. This study evaluated 168 women with gestational diabetes using blinded CGM (Medtronic iPro 2). Sleep quality was assessed using the Pittsburg Sleep Quality Index (PQSI), a widely used scale that ranges from 0-21 – higher scores indicate worse sleep quality, with scores ≥5 considered to be “poor.” About two-thirds of women had poor sleep quality and each unit increase in PQSI was associated with 11% lower Time in Range.

Diabetes Therapy Highlights

Tirzepatide is the true “game changer” in type 2 diabetes: All-star panel of KOLs highlights wide-ranging benefits in glucose, bodyweight and hypoglycemia triple endpoints, time in tight range, and diabetes remission, ultimately “moving the goalpost for T2D management”

During a session on dual agonists, an impressive KOL panel discussed the wide-reaching glycemic control and weight loss benefits of superstar dual GIP/GLP-1 agonist tirzepatide, which has been the talk of the town following the stellar SURMOUNT-1 topline results published earlier this week. Based on tirzepatide’s emerging clinical profile that spans glycemic control, body weight reductions, and even early signs of cardiovascular benefit, Dr. Julio Rosenstock (University of Texas Southwestern) stated that tirzepatide is a “true treatment paradigm change in diabetes” that has “moved the needle beyond GLP-1s.” We were especially excited to hear Dr. Rosenstock discuss the potential to achieve diabetes remission with tirzepatide. Citing a 2019 epidemiological study in Scotland, Dr. Rosenstock said that greater weight loss is associated with a greater chance of achieving diabetes remission (during Q&A, he said his preference for “remission” over “cure” given the chronic nature of diabetes). Given that 43% of participants on tirzepatide 15 mg and 28% of participants on tirzepatide 10 mg achieved weight loss ≥15% in SURPASS-3, diabetes remission appears to be well within the realm of possibility. We continue to believe that there is strong potential for tirzepatide to be explored in prediabetes as well and other preventive approaches, and we look forward to the full three-year results from people with prediabetes in SURMOUNT-1. Ultimately, Dr. Rosenstock boldly summed up tirzepatide’s remarkable clinical profile stating, “Tirzepatide has moved the goalpost for type 2 diabetes management towards attaining ‘diabetes reversal or remission,’ which may no longer be the ‘impossible dream to reach the unreachable star’!”

• Dr. Tadej Battelino (University of Ljubljana) discussed results from the SURPASS-3 CGM sub-study (read out at EASD 2021 and published in The Lancet just this week), which was the first trial to use Time in Tight Range (71-140 mg/dL) as a metric to compare pooled tirzepatide 10 mg and 15 mg doses versus insulin degludec in people with type 2 diabetes (n=243). Participants in the pooled tirzepatide group spent a whopping 73% of Time in Tight Range, compared to 48% Time in Tight Range in the insulin degludec group, for a highly significant 25% estimated treatment difference. Importantly, no significant differences were observed based on SGLT-2 inhibitor use at baseline, suggesting that these two important anti-hyperglycemic medications can be used together to improve glycemic control. In addition, Prof. Battelino highlighted the patient-reported quality of life improvements with tirzepatide relative to insulin degludec, including significantly greater improvements in total Ability to Perform Physical Activities of Daily Living (APPADL) as well as Diabetes Treatment Satisfaction Questionnaire scores (DTSQ).
• Using an impressive array of data from the SURPASS trials, Dr. Pratik Choudhary (King’s College London) identified the potential effect of the GIP/GLP-1 dual agonist tirzepatide on triple-endpoint outcomes. The data suggest that patients who are able to achieve the metabolic targets of glycemia, LDL-cholesterol, and hypertension have significantly improved microvascular and macrovascular outcomes than achieving any one metabolic target alone. Unfortunately, the number of patients actually meeting multiple metabolic treatment goals on multiple fronts is low (<25%). Data from SURPASS suggest that the potency of tirzepatide may help patients move beyond metabolic targets to a new triple endpoint: glucose (A1c), body weight, and hypoglycemia. The results from his meta-analysis show that a statistically significant greater number of participants are able to revert to normoglycemia without weight gain and hypoglycemia on tirzepatide than patients in the control groups. Dr. Choudhary called these results “absolutely astonishing.” We are greatly looking forward to the SURPASS-CVOT (expected to complete in 2024), which will offer greater insight into how achieving triple endpoints might translate into improved macrovascular outcomes.

Percentage of patients achieving A1c <5.7%

	SURPASS-1	SURPASS-2	SURPASS-3	SURPASS-4	SURPASS-5
5 mg	34	29	26	23	26
10 mg	31	45	39	33	48
15 mg	52	51	48	43	62
Comparator	1 (placebo)	20 (semaglutide)	5 (degludec)	3 (glargine)	3 (placebo)

Percentage of patients achieving A1c <5.7% without weight gain and hypoglycemia

	SURPASS-1	SURPASS-2	SURPASS-3	SURPASS-4	SURPASS-5
5 mg	32	28	25	21	23
10 mg	30	43	38	31	39
15 mg	51	50	48	42	56

• Dr. Battelino presented 24-hour glucose profiles for several case studies to illustrate the extent to which tirzepatide effectively normalizes glycemic control. Our associates weren’t the only ones gasping in disbelief when he shared these results! Ultimately, the 24-hour profiles speak for themselves, showing a clear flattening of glycemic control after one year on tirzepatide vs. persistent highs and lows with insulin degludec – that’s Flat, Narrow, and In Range (FNIR) for sure! We were especially impressed by the tremendous reductions in A1c and the very low risk of hypoglycemia compared to insulin degludec.
  
  • The first patient, a 59-year-old man, had an A1c of 8.9%, BMI of 27, and spent 5% of Time in Tight Range at the study onset on metformin alone. After one year on tirzepatide 5 mg, his ambulatory glucose profile was much flatter, with the majority of glucose values concentrated around 100 mg/dL, 95% Time in Tight Range, and an A1c of 5.3%. This patient also lost 58 kg to achieve a BMI of 19 and reported no hypoglycemia.

• • The second patient was a 61-year-old man with an A1c of 10.1% and a BMI of 41, who spent just 0.9% of Time in Tight Range at study onset on metformin monotherapy, making him a prime candidate for therapy intensification. After one year on tirzepatide 15 mg, he was spending 98% of Time in Tight Range with a virtually flat 24-hour glucose profile with all glucose values concentrated right around 100 mg/dL. We were shocked to learn that this patient’s A1c dropped a whopping 5.3% from 10.1% to 4.8% with only one episode of hypoglycemia. He also lost 86 kg to achieve a BMI 32.

• • The final case was a 74-year-old woman with an A1c of 8.7% and a BMI of 27 who spent 15% of her Time in Tight Range on metformin monotherapy. After randomization to insulin degludec for one year, she achieved a 1.7% reduction in A1c to 7.0% and was spending 63% of Time in Tight Range, with no significant changes in her weight or BMI – a stark contrast to those randomized to tirzepatide, although still a substantial improvement. Although these glycemic metrics improved, concerningly, this patient also experienced 30 episodes of hypoglycemia.

ATTD 2022 Opening Ceremony Keynote: Prof. Stefano Del Prato discusses the evolving trajectory of insulin, highlighting potential of novel oral delivery systems to eliminate the need for injection therapy

In the standing-room only Opening Ceremony, the esteemed Prof. Stefano Del Prato (University of Pisa School of Medicine, Italy) delivered an exhilarating keynote on the trajectory of insulin, tracing its evolution from a syringe-and-vial therapy to novel administration methods currently under development, like demonstration via ingestible capsules with “smart” technology to track dosing. Overall, he called insulin the “key” to opening up new avenues of treatment for people with diabetes, discussing weekly insulin formulations, oral insulins, and the use of insulin in AID systems. On weekly inuslin formulations, Prof. Del Prato highlighted basal insulin Fc (BIF) and insulin icodec, both of which have demonstrated glycemic control profiles comparable to those of once-daily insulins in phase 2 clinical trials (ENDO 2021 and ADA 2020, respectively). Turning to oral insulins, Prof. Del Prato argued that the development of inhaled insulin in 1925 – a mere five years after insulin was first discovered – paved the way for further exploration of novel administration routes. One potential oral delivery option is via intestinal micropatches, which are a pH-sensitive capsule that adhere to the mucosal layer. In a 28-day study by Eldor et al. (published in Diabetes, Obesity and Metabolism in July 2021), the oral insulin prevented increases in nighttime glucose, 24-hour glucose, and A1c without increasing the risk of hypoglycemia or other safety events compared with the control arm. Another novel administration route involves an ingestible self-orienting delivery system, in which capsules containing a needle loaded with insulin are ingested and position themselves in the mucosa. The acidic environment of the stomach degrades the needle cap, allowing for the release of insulin into the circulation. An in vivo study (published in Science in 2019) found that the self-orienting delivery system was able to achieve insulin concentrations comparable to those obtained via subcuntaneous injections. Prof. Del Prato also highlighted the potential of hybrid closed loop systems to reduce the burden of diabetes management via AI-enabled decision-making, emphasizing the algorithm’s ability to adapt to a patient’s changing insulin needs over time. Discussing the ultimate dream of an artificial pancreas, he described a prototype system that combines an insulin pump implanted in the peritoneal region with capsules that are ingested to refill the insulin pump, called PILLSID. Overall, Prof. Del Prato emphasized that insulin has achieved numerous milestones in its 100-year trajectory as the first hormone to: (i) be isolated and used for hormonal therapy in humans; (ii) be ethically distributed for production; (iii) be purified to reduce allergic reactions; (iv) be synthesized via the recombinant DNA technique; (v) have its amino acid chain manipulated to modify pharmacokinetics and pharmacodynamics; (vi) be administered on an automatic glucose feedback; and (vii) become smart (via smartpens, glucose-responsive insulins, and use in hybrid closed loop systems).

Dr. Christophe De Block suggests that once-weekly insulin would particularly benefit people not meeting glycemic targets, those who struggle to take daily insulin, and people with a consistent lifestyle; panel discussion raises questions about once-weekly insulin use during menstruation and pregnancy

During a Novo Nordisk-sponsored session, Dr. Christophe De Block (University of Antwerp, Belgium) identified who would especially benefit from once-weekly insulins and highlighted how once-weekly insulin could significantly reduce treatment burden. Of note, amidst ATTD 2022, Novo Nordisk announced topline results from once-weekly insulin icodec’s phase 3 ONWARDS 2 study, finding that icodec demonstrated a superior A1c reduction to once-daily insulin degludec. These were the first phase 3 data for any once-weekly insulin, and the potential availability of a once-weekly insulin in the next few years has raised discussion on who it would most benefit and how. Given that context, Dr. De Block identified four groups that would especially benefit from taking a once-weekly insulin: (i) people who are on GLP-1s but are not reaching their glycemic target; (ii) people with a very regular lifestyle, including those who exercise on a daily basis; (iii) those who struggle to take insulin on a daily schedule; and (iv) people in nursing homes, rehabilitation centers, and assisted living facilities. He mentioned that once-weekly insulin would not be appropriate for children, perhaps because they have such variable lifestyles.

• Responding to a question on how patients may react to injecting so much insulin at once, Dr. De Block said that it will be essential to ensure that people are comfortable with administrating once-weekly insulin and can properly recalculate their boluses. He also noted that it is still unclear how people on a once-weekly insulin should be managed if they are hospitalized. Dr. De Block also mentioned that he used to be concerned that patients would forget to take a once-weekly therapy, but his concerns were assuaged by the observation that patients on once-weekly GLP-1s have higher rates of consistent use of therapy than those on once-daily GLP-1s. He added that reminders and connected pens can resolve issues of forgetting to take therapy once-a-week.

• Following Dr. De Block’s presentation, he joined Dr. Julia Mader (University of Graz, Austria) and Dr. Richard Bergenstal (International Diabetes Center, MN) to discuss the use of once-weekly insulins in clinical practice. Dr. Mader, who is involved in the ONWARDS 6 trial investigating icodec in people with type 1 diabetes, recalled a few patient experiences from the ONWARDS 6 trial. She said one patient who often forgot to inject his insulin found that once weekly insulin was much easier to use and helped him achieve his treatment targets. Another patient who led a very stressful life found that once weekly insulin helped ease the stress of diabetes management. She also suggested that once weekly insulin may be particularly helpful for people with eating disorders. An audience member asked how once-weekly insulin use would differ during menstrual cycles, and Dr. Bergenstal said that he didn’t have an answer but that is an important area for further study. Another audience member asked how pregnancy may impact the use of insulin icodec, and Dr. Mader noted that once-weekly insulin may not be ideal in pregnant individuals since they have changing insulin requirements throughout the course of pregnancy. However, all the speakers noted that studies of once-weekly insulin in pregnancy are warranted.
• Dr. De Block cited eight factors that influence patients’ desire to initiate and consistently take insulin: (i) fear of injection; (ii) comprehension of therapy regimen/health literacy; (iii) regimen complexity; (iv) medication costs; (v) quality of communication with healthcare providers; (vi) side effects, namely hypoglycemia and weight gain: (vii) emotional well-being; and (viii) perception of benefits. He cited a 2012 survey published in Diabetic Medicine showing that both patients and clinicians identify daily injections at a prescribed time as a major burden of diabetes. The survey also found that about one-third of patients reported skipping their insulin dose at least once in the past month and that over 90% of patients wanted a non-daily insulin injection option that could help them meet their target glycemic control. Given these findings, Dr. De Block suggested that once-weekly insulin could reduce the burden for patients and caregivers, improve self-management, improve health-related quality of life,  and improve consistency of medication use.

Drs. Andrej Janež and Juan Pablo Frias look towards a new paradigm of glycemic management with once weekly insulin icodec and once weekly BIF

On the heels of the announcement of positive topline results in the ONWARDS 2 trial for insulin icodec, Drs. Andrej Janež (University Medical Center Ljubljana, Slovenia) and Juan Pablo Frias (Velocity Clinical Research, Los Angeles) reviewed once-weekly insulin analogues in clinical development. Dr. Janež discussed Novo Nordisk’s results from ONWARDS 2, in which insulin icodec led to a notable 0.22% greater A1c reduction than insulin degludec with no severe or clinical hypoglycemia. Beyond the strong phase 3 data, Dr. Janež highlighted the phase 2 icodec titration and icodec switch trials (presented at EASD 2021), both of which may inform safe and effective initiation of this novel agent. On the titration study, Dr. Janež emphasized that a less stringent titration approach (targeting pre-prandial glucose of 70-180 mg/dL with an adjustment of ± 21 U/week) enabled patients to reach their glycemic goals in a safe, effective manner. Likewise, the icodec switch study found that a loading dose (an initial double dose) is important when transitioning patients from once daily or basal insulin degludec therapy to once weekly insulin icodec. Safety wise, clinically significant and severe hypoglycemia episodes were low for all treatment groups in these studies, including the most recent ONWARDS 2 trial. Indeed, a post-hoc analysis of double-blinded Dexcom G6 data from the icodec switch study revealed that the overall median duration of hypoglycemia was similar for patients on icodec vs. those on glargine. Given the solid safety and efficacy data for icodec, Dr. Janež expressed his belief that the combination of insulin icodec and GLP-1 will offer strong glycemic control in the future – we, too, are eagerly awaiting this paradigm shift, given the greater convenience that once weekly insulin analogues afford to patients. Turning to Lilly’s once weekly insulin analog in development, Dr. Frias noted that the phase 3 QWINT program assessing basal insulin Fc (BIF) is now underway. In Lilly’s 1Q22 update, we learned that the first phase 3a QWINT trial began dosing patients last month. The 78-week trial is evaluating BIF in 939 patients with type 2 diabetes and is expected to complete in 2024. Overall, Dr. Frias emphasized BIF’s low risk of hypoglycemia and once-weekly dosing schedule, noting that once-weekly basal insulins have the potential to reduce therapeutic inertia and improve adherence and persistence with insulin therapy, thereby improving long-term outcomes. Across all of these novel insulin studies, we are pleased to see incorporation of CGM to assess Time in Range and we look forward to delving further into the CGM data for icodec and BIF as it becomes available. Ultimately, we’re pleased to see so much investment in these novel insulins that provide greater convenience and reduce the treatment burden for those on multiple daily injections.

Lilly Symposium: Tempo Smart Button CE-Marking and FDA approval anticipated in 2H22; Lyumjev “insulin evolution, not revolution” offers important marginal benefits over traditional rapid acting insulin that translate into clinically meaningful improvements in PROs

Dr. Partha Kar (Portsmouth Hospitals NHS) moderated a session with Dr. Pratik Choudhary (Kings College London) and Dr. Andreas Liebl (Fachklinik, Bad Heibrunn, Germany) reviewing the noted benefits of Lyumjev (ultra-rapid insulin lispro) and the Tempo smart button. Much of the session focused on the marginal benefits offered by these new products that enhance and improve the lives of patients with diabetes through iterative improvements on previously best-in-class products. Dr. Kar made the analogy of newer insulins to newer iPhones, stating that while an older iPhone still gets the job of answering phone calls, the newer versions are faster, more efficient, and smarter, all of which result in a happier customer. Similarly, newer insulins like Lyumjev are faster and smarter while still producing the primary result: better glucose control with a lower risk of hypoglycemia. Data from the PRONTO studies suggest that although there are no major differences in A1c at 26 weeks (see below for a more in-depth overview), patients on Lyumjev saw faster onset of insulin to reduce postprandial glucose excursions, slight reductions in hypoglycemia, +44 minutes/day Time in Range, and increased flexibility in insulin dosing. Ultimately, Dr. Liebl referred to these small improvements as an “insulin evolution, not revolution” and posited that “nearly every person with [MDI-dependent diabetes] would benefit from Lyumjev.”

• Dr. Liebl spent significant time reviewing results from the PRONTO-T1D and PRONTO-TD2 studies, which showed improved outcomes for patients on Lyumjev compared to Humalog. The PRONTO-T2D study found that ultra-rapid-acting Lyumjev was superior to Humalog in controlling one- and two-hour post-prandial glucose excursions during a mixed-meal-test through 26 weeks. Lyumjev was also, again, non-inferior to Humalog on A1c at the end of the study, with the former group reaching an average A1c of 6.92% vs. 6.86% with the latter – both meeting the target A1c goal. While the overall rates of documented and nocturnal hypoglycemia (<54 mg/dL) were nearly identical, Lyumjev conferred significantly higher rates of postprandial hypoglycemia between one to two hours after a meal (0.7 events per patient year vs 0.3, p<0.001) and between two to four hours post-meal (1.0 events per patient year vs. 0.7, p=0.04). The PRONTO-T1D study found that at the end of 26 weeks, mealtime Lyumjev demonstrated a non-inferior A1c reduction vs. Humalog (estimated treatment difference: -0.08%, 95% CI: -0.16-0.00). Moreover, while there was no significant difference in postprandial severe hypoglycemia (<54 mg/dL) between the three groups up to four hours post-meal, mealtime Lyumjev did significantly reduce these events past the four-hour mark (2.72 events per patient year vs. 4.35 with mealtime Humalog (p=0.001) vs. 3.88 with post-meal Lyumjev (p=0.006), which has been attributed to the faster onset and offset of Lyumjev. Based on these positive results, Lyumjev was approved by the FDA in July 2020 for use in adults with type 1 and type 2 diabetes and more recently, was approved for use in pumps in August 2021 based on additional data from the from the PRONTO-Pump-2 trial.
• Importantly, based on data from dQ&A, we can see that these small differences in the data are being translated into improved quality of life for patients living with diabetes. In particular, on metrics that panel participants consistently rank as the most important for mealtime insulin – consistency, amount of glucose control provided, and time to onset – Lyumjev significantly outperforms Humalog, the comparator in the PRONTO trials, as well as Fiasp, the other available ultra-rapid acting insulin on the market. It is worth noting that between 2Q21 and 4Q21, patient satisfaction for Lyumjev increased from 47% to 58%, suggesting an upward trajectory in improved patient outcomes. See the table below for a full breakdown of patient satisfaction, where the percent indicates how many respondents checked “9” or “10” on a 10-point scale (10 is the best).

	Lyumjev (n=30)	Fiasp (n=151)	Humalog U-100 (n=1,332)
Overall satisfaction?	47%	40%	44%
How much glucose control it provides?	67%	40%	36%
How quickly it works?	70%	44%	29%
How consistently it works?	57%	39%	39%
How much hypoglycemia in causes?	50%	28%	26%

• While Lilly did discuss its Tempo Smart Button, the symposium and Q&A offered no updates on system’s launch timing or a clarification on why the system was not CE-Marked by the end of 2021. According to comments from Lilly in May 2021. the company was expecting CE-Mark for the Tempo Smart Button “later in 2021,” although during today’s presentation we saw footnotes on certain slides indicating that the company has indeed not yet received CE-Marking for Tempo. That said, at the ATTD 2022 exhibit hall, we learned that the company is expecting FDA and CE-Mark to both clear by the “end of 2022,” likely at some point in 2H22 (between “summer” and “end of year”), and that the company is planning more trials to evaluate the system that will begin later in the year after the approvals. While we wish we could have learned why Tempo was not CE-Marked earlier, we are sincerely thankful for the exhibit hall representatives and hope that Lilly can navigate the challenges at the FDA and the new MDR rules to secure FDA and CE-Mark approvals by the end of 2022.
• As we learned at EASD 2021, Lilly’s Tempo system, which consists of a Tempo Smart Button, Tempo Pen, and compatible apps, builds on the company’s existing KwikPen infrastructure and stores and transmits data on patient’s insulin dose, timing, and type of insulin delivered. When launched, Lilly’s Tempo smart button will be compatible with Tempo Pens for Abasaglar 100 units/mL, Humalog 100 units/mL, and Lyumjev 100 units/mL. Lilly has also already secured partnerships with Roche, Glooko, myDiabby, and Dexcom to integrate Tempo data across these platforms, giving patients multiple options for how they may choose to view their Tempo pen data.

Bayer symposium: Dr. Antonio Ceriello highlights finerenone’s unique role in reducing CKD progression and CV complications, and Dr. Jay Skyler presents an algorithm to manage hyperkalemia associated with finerenone

Drs. Antonio Ceriello (IRCCS MultiMedica, Italy) and Jay Skyler (University of Miami) emphasized the important role of non-steroidal mineralocorticoid receptor antagonist (MRA) finerenone in treating people with CKD and type 2 diabetes. Dr. Ceriello commented on the potential mechanisms through which CKD is linked to CVD, and he highlighted the current hypothesis that finerenone targets this association by mitigating inflammation and fibrosis by adjusting gene expression. Building on Dr. Ceriello’s talk, Dr. Skyler dove into practical and clinical considerations in treating CKD in patients with type 2 diabetes and prescribing finerenone. He emphasized that, though many clinicians measure eGFR, it is also important to monitor urine albumin-to-creatinine ratio (UACR) as eGFR and UACR are independent predictors of CV mortality. Finerenone, like all MRAs, is associated with an increased risk of hyperkalemia, so Dr. Skyler shared the algorithm clinicians used in the FIDELIO-DKD and FIGARO-DKD trials to manage finerenone-associated hyperkalemia. This algorithm is of high clinical relevance as both FIDELIO and FIGARO found that even though finerenone was associated with increased hyperkalemia, there was a low rate of discontinuation due to hyperkalemia, underlining the benefits of following this treatment algorithm. 

• Dr. Ceriello discussed the mechanism by which finerenone is hypothesized to reduce CKD progression and CV complications. He explained that finerenone reduces inflammation and fibrosis by mitigating MR overactivation, which when unmitigated results in the transcription of pro-inflammatory and pro-fibrotic genes.  whereas ACEIs, ARBs, SGLT-2s, GLP-1s, metformin, and other antihypertensive medications target hemodynamic and metabolic pathways to mitigate CKD progression. This reduction in CKD progression leads to reduced CV complications, and Dr. Ceriello explained this connection by turning to the famous Steno Hypothesis published in 1989. The Steno Hypothesis suggests that albuminuria not only indicates renal disease but also reflects widespread vascular damage. In this sense, Dr. Ceriello said, “the kidney becomes a window to the vasculature.”

• • Dr. Ceriello highlighted the unique differences, characteristics and benefits of finerenone compared to steroidal MRAs. We have previously heard this topic discussed extensively at ASN 2021. Specifically, he highlighted that finerenone has high potency and selectivity for the MR with minimal sexual side effects and lower hyperkalemia compared to steroidal MRAs spironolactone and eplerenone. 

	Structural properties	Potency	Selectivity	Metabolites	Tissue distribution	Sexual Side Effects	Hyperkalemia	Effect on SBP
Finerenone	Bulky structure (nonsteroidal)	High	High	No active metabolites	Balanced in heart and kidney	Rare	moderate	moderate
Spironolactone	Flat structure (steroidal)	High	Low	No active metabolites	Higher in kidney	Gynecomastia, dysmenorrhea, and impotence	High	High
Eplerenone	Flat structure (steroidal)	Low	Medium	Multiple active metabolites	Higher in kidney	Gynecomastia, dysmenorrhea, and impotence (less common that spironolactone)	High	High

• Dr. Skyler presented a clinical algorithm to manage hyperkalemia in people taking finerenone. The FIDELIO-DKD and FIGARO-DKD trials found that finerenone is associated with an increased rate of hyperkalemia, though there was a low rate of treatment discontinuation due to hyperkalemia. This low rate of discontinuation is in part due to the trials’ algorithm to management hyperkalemia (shown below), which called for serum potassium monitoring at one month following treatment initiation and every four months afterwards. Based on the potassium level, clinicians could raise the finerenone dose up to 20 mg, maintain the current finerenone dose, or withhold finerenone temporarily.  

• • Dr. Skyler commented on finerenone’s use in adolescents, with SGLT-2s, and in people with type 1 diabetes. During Q&A, in response to a question about finerenone use in adolescents with type 2 diabetes, Dr. Skyler said that he would consider off-label use but would like to see a study in this population. Following the presentation, our team spoke with Dr. Skyler about finerenone use with SGLT-2s and in people with type 1 diabetes. He noted that studies are currently being designed to investigate both applications of finerenone and that it is too early to speculate on finerenone’s benefits in people with type 1 diabetes.

Arecor’s rapid-acting concentrated insulin aspart U500 (AT278) demonstrates enhanced early insulin exposure and accelerated glucose-lowering compared to NovoLog in PK/PD study

In this oral abstract presentation, Dr. Eva Svehlikova (Medical University of Graz, Austria) presented results from a phase 1 study (n=38) investigating Arecor’s rapid-acting concentrated insulin aspart U500 (AT278) compared to Novo Nordisk’s NovoLog (insulin aspart U100). AT278 demonstrated an earlier insulin exposure and glucose-lowering effect than NovoLog. Specifically, AT278 led to a four-fold higher insulin exposure within the first 30 minutes of injection and a two-fold higher glucose-lowering effect within the first 60 minutes of injection. AT278 became present in the blood stream six minutes earlier and had a 10-minute earlier onset of action than NovoLog. Importantly, the overall insulin exposure and glucose-lowering effect of AT278 was similar to NovoLog, and AT278 was well-tolerated with no safety signals detected. Dr. Svehlikova concluded that AT278 maintains the rapid-acting insulin characteristics of NovoLog but in a reduced volume and has the potential to improve blood glucose management and convenience for people with high insulin needs. She added AT278 also has the potential to match the demands of next generation insulin delivery devices with smaller reservoirs.

Big Picture Highlights

Inaugural ATTD press conference with Drs. Tadej Battelino and Moshe Phillip; themes of technology for type 2s, expanding access to technology, and driving forward innovation

ATTD 2022 conference organizers Dr. Tadej Battelino (University Medical Center Ljubljana, Slovenia) and Dr. Moshe Phillip (Schneider Children’s Medical Center, Israel) joined our very own Ms. Kelly Close (Close Concerns) to host the first ever ATTD press conference with members of the media to discuss upcoming sessions and themes of the conference. Gathered in one of the smaller meeting rooms, the session served as an opportunity to attendees to ask questions directly to Drs. Phillip and Battelino who then discussed the topics of advancing technology use among people with type 2 diabetes, expanding access to technologies globally, and the role of innovation in the advancement of diabetes technologies and therapeutics.

• Discussing the applications on diabetes technology for people with type 2 diabetes, Dr. Battelino highlighted the recently established consensus group on the use of CGM among type 2s to intensify therapy and Dr. Phillip emphasized the benefit of decision support systems for providers caring for people with type 2 diabetes. Dr. Battelino expressed strong support for increased CGM use among people with type 2 diabetes, especially those taking insulin, but further emphasized that CGM data is beneficial for the optimization of any therapeutic agent – here Dr. Phillip reminded attendees that CGM can also help drive substantial behavior change as was seen in the MOBILE study. Additionally, Dr. Battelino expressed strong support for potential CGM applications related to screening for prediabetes and raised the question of whether or not CGM could be a useful tool in population-level screening for type 1 diabetes. On decision support, Dr. Phillip discussed the benefit of these AI-driven systems to help providers deliver the highest quality of care to their patients, especially providers in primary care who may not be as knowledgeable about best practices in diabetes management, but who are treating the majority of people with type 2 diabetes. Expanding uptake of diabetes technology to people with type 2 diabetes has been a strong theme thus far at ATTD 2022 with poster presentations on the feasibility and human factors data from people with type 2 diabetes using Insulet’s Omnipod 5 AID system as well as sessions CGM use among non-insulin treated type 2s. Looking ahead to the rest of ATTD 2022, we expect this to be a continuing theme with what is expected to be an extremely impactful set of presentations from Drs. Rich Bergenstal (International Diabetes Center) and Dr. Thomas Martens (International Diabetes Center) on a novel systematic approach to using CGM to direct therapeutic treatment in patients with type 2 diabetes.
• Following a question from an IDF representative, Drs. Phillip and Battelino outlined strategies to improve access to technologies in low-and-middle-income countries (LMICs). Specifically, Dr. Phillip highlighted the importance of education championing hybrid scientific meetings such as ATTD 2022 as opportunities for clinicians and scientific leaders who may not be able to travel to in-person meetings to access learnings about technology adoption and implementation. Additionally, Dr. Phillip discussed the cost of diabetes technology advocating for programs that can reduce the sometimes-prohibitive costs of technology to help drive increased adoption. Similarly, Dr. Battelino shared his view that as diabetes technology continues to advance, the baseline level of technology available to more people will increase, so while he recognized that not everyone with diabetes may be able to access the “Ferrari version” of diabetes technology, he believe we will soon be in a place where technology that is very capable of driving improvements in glycemic management and outcomes will be more broadly available. Ms. Close also joined in on this conversation to suggest that some level of diabetes technology use, whether intermittent CGM or pump therapy could be extremely beneficial to help patients and providers in LMICs gain a better sense of each individual’s glycemic profile and help tailor future treatment without requiring consistent technology use. For example, intermittent CGM can help identify necessary changes in insulin therapy, or short-term pump or AID use could help patient and providers better identify an ideal total daily insulin dose.
• Highlighting the innovation and incredible science being presented at ATTD 2022, Drs. Battelino and Phillip reflected on the benefits of being able to return to in-person meetings and made special mention of the meeting’s many start-up participants. Specifically, ATTD 2022 has 23 diabetes start-up exhibitors, a number of which received financial assistance to attend the meeting and share their technology with attendees. Both Dr. Battelino and Dr. Phillip commended the great work of many of these start-up leaders and encouraged in-person attendees to visit the start-up section of the exhibit hall to take advantage of the “collaborations and communications” that are more easily possible with in-person events. Additionally, Dr. Phillip shared his view that while virtual meetings have been very successful over the last two years, he finds that the more causal conversations that can take place informally at in-person meetings are often where some of the best ideas originate and he expressed hope that ATTD 2022 would be able to foster this type of collective learning to drive forward innovation in diabetes technologies and therapeutics.

24% agreement between sensor-detected and patient-reported hypoglycemic events (<70 mg/dL) among Hypo-METRICs participants over 10 weeks; 33% agreement rate for severe hypoglycemic events (<54 mg/dL)

Dr. Patrick Divilly (King’s College London) presented data on the rates of sensor and participant detected hypoglycemia in the Hypo-METRICS trial finding only 24% of sensor detected hypoglycemia events were matched with patient reported hypoglycemia. Specifically, in the Hypo-METRICS trial, participants (n=312) wore blinded FreeStyle Libre 2 CGMs and used the Hypo-METRICS app for 10 weeks to identify and record sensor-detected (CGM readings <70 mg/dL for at least 15 minutes) and participant-reported hypoglycemia (symptomatic hypoglycemia event resolved with carbohydrate ingestion or BGM <70 mg/dL), respectively. Participants had an average age of 56, had either type 1 (n=178) or type 2 (n=134) diabetes, 58% used MDI, 24% were on pump therapy, and 60% used personal CGM. Of note, 23% of participants had hypoglycemia unawareness as identified based on GOLD questionnaire scores. Over the 10-week study duration, there were a total of 8,880 patient-reported hypoglycemic events compared to a total of 19,335 sensor-detected hypoglycemic events (<70 mg/dL). Analysis was then conducted to match patient-reported and sensor-detected hypoglycemic events with a matched pair defined as those with a patient-reported event within +/- 1 hour of the sensor-detected hypoglycemia. Notably, only 24% of sensor detected hypoglycemia <70 mg/dL matched with a patient-reported hypoglycemic event. Interestingly, this agreement rate differed in people with type 1 versus type 2 diabetes with 30% of sensor detected hypoglycemia in patients with type 1 diabetes matching a patient-reported hypoglycemic event, compared to 15% of sensor detected events among patients with type 2 diabetes. These data suggest that there are a substantial number of sensor-detected hypoglycemic events that may go undetected by patients. As these data did not specify the glycemic level for each hypoglycemic event, we are curious if there may have been a large number of events during which patients experienced glucose levels of ~70 mg/dL (i.e., close to in range values), and thus not experienced symptomatic hypoglycemia. In addition to sensor-detected hypoglycemia that did not match with patient-reported events, there were also 12% of patient-reported hypoglycemic events that did not match with any sensor-reported events potentially suggesting that as patients began to experience hypoglycemic symptoms they took corrective actions and were able to prevent a sensor-detected hypoglycemic episode.

• Dr. Divilly also broke out sensor-detected and patient-reported severe hypoglycemic events (<54 mg/dL) which had a similarly low match rate at 33%. Again, this rate differed by type of diabetes with 33% of sensor-detected severe hypoglycemic events among type 1s matching with a patient-reported event whereas only 15% of sensor-detected severe hypoglycemic episodes matching patient-reported events among type 2s. In total, there were 5,338 sensor-detected severe hypoglycemic events with 4,343 among patients with type 1 and 995 among patients with type 2.

“Not all hypoglycemia is equal”: Drs. Stephanie Amiel and Pratik Choudhary call for research to develop standardized definition of hypoglycemia and to better understand its impact on daily functioning

Leading experts in the field of hypoglycemia research, Dr. Stephanie Amiel (King’s College London) and Dr. Pratik Choudhary (University of Leicester, UK) called for further research on hypoglycemia, both symptomatic and asymptomatic, to develop a data-driven definition and to better understand its impact on various domains of patients’ daily lives. Dr. Amiel opened with a provocative question: “It’s 2022. Why are we still talking about hypoglycemia?” She argued that residual severe hypoglycemia, impaired hypoglycemia awareness (IAH), and the imperfections in diabetes technology access, engagement, and capabilities necessitate ongoing work to address this acute complication of diabetes.

• Dr. Amiel said that IAH is the top modifiable risk factor for severe hypoglycemia and can be addressed by structured education and interventions. According to an excellent presentation by Dr. William Polonsky on restoring IAH, it is crucial that interventions not only help patients be safe but also help patients feel safe. Such interventions can include relaxation training, practice correcting via carb intake, and gradual behavior exposure. On interventions for IAH, Dr. Amiel referenced the HARPdoc study, which uses motivational interviewing and cognitive behavioral therapy to address harmful cognitions surrounding hypoglycemia in those with IAH. Initial results found that the HARPdoc intervention was associated with substantial reductions in the number of severe hypoglycemia events, while the latest data presented at ATTD 2022 Day #2 and published in Nature concluded that the intervention led to significant reductions in diabetes distress, anxiety, and depression that were sustained throughout the trial.
• Dr. Choudhary discussed the HypoMETRICS study, which seeks to bridge the knowledge gap in hypoglycemia by developing a standardized definition of this diabetes complication. More specifically, HypoMETRICS aims to determine optimum parameters of sensor-detected hypoglycemia that best correlate with events that patients experience as hypoglycemia. Dr. Choudhary argued that current definitions of hypoglycemia, such as those from the International Hypoglycemia Study Group, tend to be “one dimensional” because they are based solely on glucose thresholds and fail to consider hypoglycemia duration. HypoMETRICS is recruiting a total of 600 people, 200 of whom have type 1 diabetes with intact hypoglycemia awareness, 50 of whom have type 1 diabetes with IAH, and 350 people with type 2 diabetes on multiple daily injections (2+ injections/day). Preliminary data collected from 312 participants has documented 21,000 hypoglycemia events with substantial overlap between symptomatic and asymptomatic hypoglycemia. Ultimately, the trial will help determine the impact of hypoglycemia on daily functioning in multiple domains of life in order to develop a data-driven definition of hypoglycemia. As we eagerly await the forthcoming consensus guidelines on CGM use in clinical trials, we look forward to learning about updated definitions of hypoglycemia from the HypoMETRICS trial that may better serve patients, providers, and researchers. We especially appreciate the trial’s focus on asymptomatic hypoglycemia and believe this is a key area where CGM can provide insight to inform diabetes management decisions.

Diabetes care is a function of metabolic and psychosocial parameters: Dr. Frank Snoek focuses diabetes distress while challenging the notion that “technology will set you free”

Kicking off the second half of ATTD 2022’s Abbott School, Dr. Frank Snoek (Amsterdam University Medical Center, Netherlands) stressed the importance of using metabolic and psychosocial parameters to assess diabetes care. Dr. Snoek’s presentation echoed many KOL sentiments we observed in 2021 calling for a greater emphasis on diabetes distress and also more standardization in the collection and use of PROs and QOL metrics to inform patient-centered, personalized care. Dr. Snoek argued that diabetes management comes down to ~95% self-management, which can be quite difficult and burdensome to balance everyday given the 42 factors that affect blood glucose. On sources of emotional distress, Dr. Snoek said that: (i) diabetes is a 365 day-a-year proposition/”job”; (ii) behavioral effort does not always translate to improved outcomes (“effort-reward imbalance”); (iii) food restriction concerns; (iv) lifestyle changes are hard to do and even harder to maintain; (v) acute, unpredictable glycemic excursions are disruptive, stressful, and sometimes shameful; (vi) coping with the threat of long-term complications can be debilitating; and (vii) discrimination and “negative” support. Dr. Snoek explained that these sources of emotional distress can lead to diabetes burnout and a significant elevation of psychological comorbidities.

• Turning to technology, Dr. Snoek interrogated the impact of diabetes technology on quality of life. Dr. Snoek specifically addressed a notion brough up in Nature Communications from June 2021 that “technology will set you free,” arguing that people must be careful by saying that as a blanket assumption. As an example, Dr. Snoek highlighted the Inreda bihormonal AID system, which some might call an “artificial pancreas” given that it is an AID system that does not require user-initiated meal announcements or significant daily user input (we hesitate, however, to use this term considering that most commercial AID systems do require frequent user input and also because even bihormonal systems at this stage are not “set it and forget it”). In the system’s randomized crossover trial, the investigators found “no differences” in quality of life, psychological adaptation, and treatment satisfaction. While they note that the number of patients included in the trial might have been too small (n=23) to demonstrate any differences in this questionnaire score, they suggest that the complexity of a bihormonal system might have posed a burden to participants, which might have offset any positive QOL effects of a bihormonal AID system. Dr. Snoek also presented a table from a review of CGM studies, showing that while some studies observe QOL improvements from CGM, others fail to replicate this finding, particularly in the case of alarms overnight. That said, in his tempered analysis, Dr. Snoek also pointed out the fact that there are often measurement issues with QOL and that it’s incredibly difficult altogether for anything to meaningfully and significantly affect QOL. Likewise, he made a point to note that he did not want to negate the positive QOL outcomes that many people with diabetes continue to gain from CGM, insulin pumps, and AID.

• Diving further into how diabetes and wellbeing are related, Dr. Snoek laid out some mediating variables. On the direct and physiological side, Dr. Snoek identified: (i) Time in Range, which directly influence’s one’s mood; (ii) Time Below Range and hypoglycemia, which also influences mood; (iii) sleep, which influences vitality and cognitive function; and (iv) postprandial excursions, which also affect mood. We found Dr. Snoek’s indirect mediating variables list to be fascinating too: (i) one’s sense of control/“mastery” of diabetes care (i.e., understanding, safety); (ii) one’s sense of feeling supported; (iii) and one’s sense of perceived effort and burden.
• Pointing to a “new opportunity on the horizon,” Dr. Snoek highlighted a recently published paper by Lori Laffel et al. on “precision monitoring in diabetes.” Entitled “Coordination of glucose monitoring, self-care behavior and mental health: achieving precision monitoring in diabetes” and published in Diabetologia, the authors argue that the integration of behavioral and mental health data into the analysis of glucose data “could enrich [the] identification of subgroups to stimulate precision medicine.” Dr. Laffel et al. suggest that automatically integrating behavior, mental health and glucose data could enable the identification of certain subgroups that, for example, show a strong association between mental health and glucose, as well as those that don’t. This is quite fascinating, and seems like an extension of the glucotype subgroups concept that we’ve heard of at past conferences while also reflecting KOL sentiments from the Digital Quality Summit 2021 to include diabetes distress in the AGP report. We were also interested to see the term “just in time adaptive intervention” used as a potential application of this idea, as we currently associate that term with Glooko’s acquisition of xbird.

Future of nutrition management is bright: Innovative technology like Rocket AP, creative AI like Medtronic’s Klue, and multihormone therapy are potential solutions to help reduce burden

The esteemed Dr. Bruce Buckingham gave a compelling talk highlighting the future of nutrition management with closed loop insulin delivery. Dr. Buckingham began by framing the issue: 65% of patients miss more than one meal bolus/week, which is understandable given the realities of living with a chronic disease, but unfortunately leads to worse glycemic control. Although a truly closed loop system remains the goal and existing hybrid closed loop has reduced this burden to some extent, all available systems still require some sort of meal announcement. Late or missed meal boluses on AID tend to increase prandial hyperglycemia and can increase the risk for late post-prandial hypoglycemia, which is most common when given >1 hour after a meal or when blood sugar is >200 mg/dL. However, there are a number of algorithms in development that are intended to improve meal-time detection, including RocketAP, the algorithm under development with the UVA researchers who developed the Control-IQ algorithm. This algorithm builds on Control-IQ with a new bolus priming system module that is designed to detect unannounced meals quickly and deliver bolus insulin before a lengthy hyperglycemia episode begins. Recall that a study from UVA presented at ATTD 2021 and published in November 2021 showed highly promising results on the algorithm: RocketAP delivered +30% Time in Range over Control-IQ during the six-hour period following unannounced dinner (n=18). Dr. Buckingham also highlighted alternative innovations, like Medtronic’s Klue, a “gesture-sensing” and AI software driven by data from Apple Watch to detect when and how fast someone is eating or drinking, which could help reduce the burden of mealtime bolusing.

• Dr. Buckingham also discussed the potential of insulin co-formulations and multihormone therapy to reduce post-prandial excursions.  In particular, he highlighted a 2020 paper that found insulin+pramlintide increased time in range from 74% to 84% with a major flattening of post prandial glucose. Two additional studies with pramlintide (Majdpour Meng 2021 and Andersen 2021) confirmed these results. Interestingly, a 2020 study randomizing patients on full closed loop to dapagliflozin or placebo found that dapagliflozin stabilized nighttime glucose excursions and significantly increased Time in Range following two unannounced meals.
• We can’t help but wonder if there is greater opportunity with ultra-rapid acting insulins like Lyumjev to help reduce the burden of late boluses, particularly given that Lyumjev, approved for use by the FDA in July 2020, carries an additional indication for post-meal dosing that allows for bolusing in the first 20 minutes of the meal. The greater flexibility to dose during mealtime leads to improvements in post-prandial glucose excursions compared to other rapid acting insulins.

Preliminary T1DEXI results suggest that aerobic exercise leads to biggest drop in blood glucose; structured exercise can lead to clinically meaningful improvements in Time in Range

Dr. Michael Riddell (York University) presented a first look at the results from the T1DEXI study to a standing-room only audience. The T1DEXI study was an “at home” observational study of exercise and type 1 diabetes that aimed to: (i) better understand how different forms of exercise influence glycemia and acute glycemia in people living with T1D; (ii) determine key individual and/or event variables that influence glycemic control during exercise; and (iii) improve current exercise guidelines for type 1 diabetes. The four-week long study randomized 561 participants to one of three exercise protocol groups: (i) aerobic exercise (n=162) (ii) high intensity interval exercise (n=165); and (iii) resistance exercise (n=170). Participants followed “at home” exercise videos designed to help participants achieve a target heart rate (age-dependent, differed based on treatment group) and lasted for 25-30 minutes. To collect data, participants used body sensors and wearables to track glucose and activity, and logged physical activity, carb intake, and insulin use on the T1DEXI app, created by Dr. Pete Jacobs at OHSU. To learn more about the study design and goals, read our interview with T1DEXI study founders from October 2020 here. Ultimately, this data will contribute to the collective knowledge of exercise in type 1 diabetes, which can be a burdensome source of confusion and anxiety for patients with the condition and we look forward to hearing more detailed results at ADA in June. Notably, this data will be housed in the open access Clinical data Interchange Standards Consortium format, so all interested parties will be able to access this data. We applaud the T1DEXI study’s commitment to making this data publicly available to support innovation and progress for improving the lives of people with diabetes.

• The participants in the T1DEXI study were primarily female (73%) and young – middle age, with a mean age of 37 years. Participants had an average A1c of 8.6% and an average diabetes duration of 18 years. The average BMI was 25.4 kg/m² and were on a mix of insulin delivery systems: 45% on closed loop, 18% on MDI, and 37% on injections. Approximately 19% of participants in the study had impaired hypoglycemia awareness and most had some baseline level of physical activity (50% minimally active, 42% HCPA active).

• The results found that, as expected, aerobic exercise led to the greatest change in blood glucose during exercise sessions, falling an average 19 mg/dL per video. Interval training videos led to a 15 mg/dL drop in blood glucose and the resistance training led to a 9 mg/dL drop in blood glucose. Thus, when exercising, people with diabetes should expect aerobic exercise > interval training > resistance training to have the largest impact on blood glucose levels. Each of these values was statistically different from one another.
• Several individual characteristics impact the magnitude of glucose change during study exercise. Beyond type of exercise, males, participants with higher baseline A1c, participants with higher baseline blood glucose levels, participants with lower time in range, and participants with lower heart rates saw greater drops in blood glucose during the exercise period. These factors should be considered when figuring out how to dose insulin for exercise.
• Notably, the T1DEXI results found that all three forms of exercise were associated with an acute improvement in Time in Range over the 24 hours following the exercise period. Also, the change in Time in Range produced by the exercise videos, which we interpreted on the graphs to be around 5-7%, is considered clinically meaningful, which is an added bonus to the effects of exercise.
• Dr. Riddell made the important point that “exercise” and “physical activity” do not necessarily mean the same thing, but both impact how the body utilizes glucose. While “exercise” is a structured form of physical activity that is performed with the intent to maintain or improve health and fitness, “physical activity” is defined as any body movement caused by the contraction of skeletal muscle that substantially increases energy expenditure compared with rest, and includes things like gardening, housework, or grocery shopping. A word cloud depicting the types of activities that people participated in featured walking, biking, jogging, pilates, housework, and, of course, the study videos.

Debate on modifying T1D progression: Dr. Desmond Schatz argues type 1 diabetes is not one disease and calls for precision medicine, Dr. Chantal Mathieu responds by highlighting common denominators of disease progression and opportunities to prevent type 1 diabetes

Drs. Desmond Schatz (University of Florida) and Chantal Mathieu (KU Leuven, Belgium) discussed paradigms for modifying type 1 diabetes progression. Through highlighting the enormous heterogeneity in type 1 diabetes over time and between individuals, Dr. Schatz argued that the type 1 field must identify disease subtypes with distinct functional or pathobiological mechanisms (endotypes) and identify additional specific biomarkers that can aid in predicting the efficacy of and evaluating treatments. He also called for more homogenous trial groups through precision medicine-directed enrollment criteria. He explained that type 1 diabetes is not one disease, and its treatment paradigm should become more like cancer, where a host of biomarkers are used to identify specific manifestations of the disease with distinct treatment approaches. Responding to Dr. Schatz, Dr. Mathieu acknowledged that there is a lot left to learn about the heterogeneity of type 1 diabetes and the field should move toward precision medicine, but she said that there can be steps along the way to a precision medicine approach. She argued that there are common denominators among people with type 1 that present opportunities to modify disease course. She highlighted lessons and developments made in immunotherapies to modify type 1 diabetes progression. 

• Dr. Schatz argued that the limited success in preventing/reversing type 1 diabetes is because it is a heterogenous disease with various mechanisms. Given this, Dr. Schatz called for the field to identify type 1 endotypes. He said that type 1 diabetes occurs at different ages and with different rates of progression and is associated with different complication rates, genetic susceptibilities,  and therapeutic efficacies. He added that the mechanism(s) of type 1 diabetes is not clear: is type 1 caused by dysfunctional beta cells being attacked by a healthy immune system or a dysfunctional immune system improperly attacking healthy beta cells? Much of Dr. Schatz’ talk was dedicated to highlighting the enormous heterogeneity in type 1 diabetes over time, among individuals, and even within an individual. Citing data from the T1D Exchange, he said that adolescent type 1 is notoriously difficult to manage compared to type 1 diabetes in adults, indicating that the disease may vary with time. He acknowledged that environmental factors certainly affect diabetes over time, but even differing environmental factors that affect disease progression are key considerations in the path to modify disease course. He also pointed out that C-peptide preservation varies with age at diagnosis and that patients of different ages tend to have different autoantibodies (adolescent typically have IAA only, whereas older adults typically have GADA only). Ultimately, he said researchers must define endotypes for type 1 diabetes, which are diabetes subtypes defined by distinct functional or pathobiological mechanism. (Endotypes denote mechanism, whereas phenotypes do not). He said that identifying endotypes that account for both genetic and environmental influences could help explain clinical heterogeneity and facilitate precision medicine.
• Dr. Mathieu argued that the field has come a long way in understanding the common stages of type 1 diabetes and that these stages have provided important lessons and opportunities for disease modification. She wholeheartedly agreed with Dr. Schatz that type 1 is a heterogenous disease and that precision medicine will be important to optimize treatment. That being said, she said that it has become clear that people with type 1 diabetes go through multiple stages that have important implications for treatment. Highlighting the importance of therapy timing based on the stages of disease progression, Dr. Mathieu said that giving insulin to young non-obese diabetes (NOD) mice before the immune system has destroyed beta cells can prevent beta cell destruction. In an attempt to translate this finding to humans, researchers gave insulin to adults in later stages of T1D and found that it wasn’t effective. She said that the timing of an intervention is crucial, citing a 1997 study showing that anti-CD3 immunotherapy was effective in modifying disease course when given to mice with stage 3 (clinical) diabetes, but not when given to mice with earlier stages of disease. Dr. Mathieu also highlighted advancements that have been made in modifying type 1 diabetes progression. She specifically raised anti-CD3 therapies teplizumab and otelixizumab and anti-TNF alpha therapy golimumab as three promising therapies that have been shown to modify type 1 disease course. She said that combining therapies, such as immunotherapies with beta-cell preserving agents, could lead to more significant improvements in disease progression. She concluded that there are common denominators in type 1 diabetes that allow short term immune modulation to temporarily arrest the loss of functional beta-cell mass, and in the long-term better biomarkers are needed to facilitate precision medicine and distinguish between responders and non-responders.
• We note that any advancements in modifying type 1 diabetes progression must be made in conjunction with advancements in screening and identifying those at risk for type 1 diabetes progression. The availability of a disease-modifying therapy, such as Provention’s teplizumab, would catalyze the conversation around population-based screening, helping place type 1 autoantibody screening as a standard part of pediatric care. For more on expanding awareness, access, and use of T1D autoantibody screening, see the Milken Institute’s and Helmsley Charitable Trust’s report “Type 1 Diabetes Autoantibody Screening: A Roadmap for Pediatric Policy Implementation.” 

Dr. Jay Skyler discusses the journey from “hype to hope” in type 1 diabetes research, emphasizing encouraging results with teplizumab, combination therapy with GLP-1s, and AID

The esteemed Dr. Jay Skyler (Diabetes Research Institute, Florida) discussed the journey to find a cure for type 1 diabetes, highlighting the progress from the first reversal of diabetes in rats in 1979 to the highly encouraging clinical trials with teplizumab today. Echoing his talk at Keystone 2021, Dr. Skyler outlined the four ideal therapeutic goals in type 1 diabetes research: (i) prevention of immune destruction; (ii) preservation of beta cell mass; (iii) replacement and regeneration of beta cells; and (iv) automated insulin delivery. Overall, Dr. Skyler argued that there is strong evidence supporting the use of therapeutic agents to prevent or delay type 1 diabetes onset or progression, but regulatory bodies have hindered progress by failing to support long-term trials that would clarify the full effects of these drugs on beta cells. On prevention of immune destruction, Dr. Skyler highlighted anti-CD3 agent teplizumab, which is the first drug to show preservation of beta cell function. Recall that teplizumab treatment led to an average three years delay in median time of diagnosis in the TN-10 study (presented at ADA 2019). In particular, Dr. Skyler emphasized teplizumab’s long-lasting effects after an initial 14-day course of treatment, highlighting that 18% of people in the teplizumab group still did not have diabetes after five years compared to just 6% of those in the placebo group. We were quite intrigued to hear Dr. Skyler discuss the use of GLP-1s in combination with other agents to further protect beta cell health. For instance, a study published in The Lancet in March 2021 found that decreases in C-peptide levels from baseline to one year were significantly smaller with combination therapy of anti-IL-21 and liraglutide compared to either drug alone (n=308). Given these results, Dr. Skyler lamented that the study only lasted 54 weeks, asking, “if you’re giving a drug to improve function, why do you want to stop it if it improves function?” Turning to beta cell replacement and regeneration, Dr. Skyler highlighted stem cell-produced islets as a key source of replacement beta cells, given the limited availability of cells (there are only 12,000 potential pancreas donors each year in the US) and lack of reimbursement for islet cell transplants in the US. To date, key milestones include Viacyte’s PEC-Direct islet therapy (presented at ADA 20121, preliminary results published December 2021) as well as the recent positive topline results from Vertex’s phase 1/2 trial of VX-880. Lastly, on AID, Dr. Skyler discussed positive results from Tandem’s Control-IQ pivotal published in NEJM back in 2019. Dr. Skyler stated that “every patient with type 1 diabetes who wants should be offered the opportunity to be on AID and we need to get our reimbursement systems around the world online to do that.” We were thrilled to hear him advocate for the use of CGM in type 1, type 2, and even pre-diabetes to help patients understand how their lifestyle, behavior, and food choices affect glycemic control. Ultimately, Dr. Skyler concluded that we have moved from a stage of “hype” to one of hope in type 1 diabetes, in which it may soon be possible to achieve the four therapeutic goals thanks to developments in both therapies and diabetes technology.   

• Dr. Moshe Phillip and Dr. Tal Oron of Schneider Children’s Medical Center of Israel discussed the Antibody Detection Israel Research (ADIR) project as a model for a national, general population screening at a young age. The five-year initiative, funded by a grant from JDRF, aims to screen 50,000 children by adding autoantibody screening onto the mandatory complete blood count that all infants receive. Dr. Phillip emphasized that Israel is an ideal setting in which to conduct this study as it is a small, densely population country with a high proportion of children, all of whom can easily be tracked because they are enrolled in one of the country’s four public health systems. For those who test positive for autoantibodies, follow-up will be provided, focusing on education about preventing DKA, recognizing the signs and symptoms of diabetes, and reducing stress for parents. Overall, Dr. Phillips characterized this community-based screening program as an ideal model for future programs in other countries and called for further action to change the paradigm of type 1 diabetes through the #WeAreNotWaiting movement.
• Prof. Thomas Danne (Diabetes-Zentrum, Hannover, Germany) discussed different approaches to type 1 diabetes screening in Germany in the FR1da and Fr1dolin programs and highlighted the POInT and SINTIA primary prevention trials for those who have elevated risk of type 1 diabetes. Both programs combine screening for autoantibodies for type 1 diabetes with other screening tests (COVID-19 in FR1da and hyperlipidemia in Fr1dolin, respectively) to make the testing more attractive to policymakers and HCPs. In interventions, Prof. Danne highlighted the POInT trial, which is investigating whether insulin powder can be used to train the immune system in 1,050 children. Prof. Danne also described the SINT1A primary prevention trial that is evaluating dietary supplementation with b. infantis to mitigate type 1 diabetes autoimmunity. Overall, Prof. Danne advocated for screening programs such as Fr1da and Fr1dolin in order to reduce the DKA rate, provide preventive education on type 1 diabetes, and to allow for a controlled transition to insulin therapy for those who test positive for multiple autoantibodies.

Dr. Thomas Danne explores seasonality of type 1 diabetes diagnoses during the COVID pandemic in new analysis from the SWEET registry

Dr. Thomas Danne (Hannover Medical School), Chairman of the global pediatric SWEET Project, used data from the 108 centers in the registry to further explore the relationship between diabetes and COVID. In September 2021, Dr. Danne and his group published an article showing that although glycemic control was maintained from baseline, rates of DKA significantly increased in countries with highest COVID mortality during the wave that occurred in May/June 2020. Additionally, recent research from Germany found a significant increase in the incidence of type 1 diabetes in children during the COVD-19 pandemic, with a delay in the peak incidence of type 1 diabetes until roughly three months after peak COVID-19 cases and the strictest pandemic containment measures. To further explore this timing of diabetes diagnoses related to COVID-19, Dr. Danne highlighted new data showing patterns of diabetes onset seasonality in 2020 in comparison to previous years. While the annual trend of growing type 1 diabetes incidence continues, the months in which people are most likely to be diagnosed have been disrupted by COVID. In most years, T1D diagnoses are the highest in cold winter months and the lowest in summer months. However, as seen in the graph below, 2020 disrupted this pattern, with lower rates of diagnoses in the winter months and the highest rates of diagnosis in the summer. Notably, however, by 2021, these rates had mostly returned to the expected pattern. Dr. Danne did not draw firm conclusions on why this might be the case but pointed out that this could be due to direct or indirect influences from the pandemic and both must be taken into consideration in future research.

• The data comes from 88 centers in Europe, Africa, Asia, Middle East, Australia, New Zealand, United States, Canada, and South and Central America. Importantly, this is a longitudinal analysis, and centers were only eligible if they had collected data from 2018-2021. The patients included in the study all had type 1 diabetes and were younger than 18 years old. Patients were excluded from the study if data on age, sex, diabetes type or diabetes duration was missing from the chart.
• At baseline, most clinical characteristics were comparable across study years, including the ratio of male and female patients being analyzed, the age at diagnosis, and annual change in BMI. Interestingly, however, the yearly A1c average was, at 8.5%, significantly greater in 2020 compared to 8.2% in 2018, 2019, and 2021 (p<0.001). Dr. Danne did not speculate on the reasoning behind this too much, but perhaps it could be inferred that delays in care that occurred as a result of the pandemic ultimately led to greater increases in A1c.
• The SWEET registry continues to be an important source of data collection for the management of type 1 diabetes. Notably, in 2021, more than 48,500 unique patients visited a SWEET center, with nearly 140,000 total visits. SWEET has adapted to telehealth visits, rising from “basically none” before the pandemic to up to more than 14,000 in 2021. Recall that the vision for SWEET is to provide equal, quality care for children with diabetes and to optimize outcomes around the world.

Nature Communications publishes HARPdoc trial, showing efficacy (but not superiority) of novel psychoeducational program to reduce severe hypo in type 1s with impaired hypo awareness (n=88); first in-person ATTD Forum (!)

Prof. Stephanie Amiel (Kings College London, UK) presented results from the HARPdoc trial to an intimate audience at the ATTD Education Forum, coinciding with a simultaneous publication in Nature Communications. The publication was entitled “A parallel randomised controlled trial of the Hypoglycaemia Awareness Restoration Programme for adults with type 1 diabetes and problematic hypoglycaemia despite optimised self-care (HARPdoc).” We first saw the results from the “Hypoglycemia Awareness Restoration Program for adults with type 1 diabetes and problematic hypoglycemia Despite Optimized Control” (HARPdoc) study at EASD 2021, where Dr. Amiel noted that although HARPdoc was associated with substantial reductions in the number of severe hypoglycemia events at 12 and 24 months, it did not achieve superior reductions to the “Blood Glucose Awareness Training” (BGAT) educational program that was developed in the late 1980s. And yet, and most impressively, HARPdoc demonstrated significantly greater reductions in diabetes distress, anxiety, and depression than BGAT, and these improvements in mental health were sustained over the entire duration of follow up. We were privileged to be part of this intimate audience, and it was so amazing to attend our first ATTD Forum in person! This is the fifth special ATTD Education Forum session that we’ve attended since ATTD 2021, during which the conference organizers held a special session for the readout of the MOBILE study, which was simultaneously published in JAMA.

Dr. William Polonsky shares practical advice for helping patients overcome fear of hypoglycemia, with emphasis on reducing risk through CGM use and restoring confidence through behavioral strategies

Dr. William Polonsky (Behavioral Diabetes Institute, San Diego, California) discussed best practices for addressing fear of hypoglycemia, focusing on the importance of restoring patients’ confidence in their ability to manage a severe hypoglycemia episode. While having some fear of hypoglycemia is healthy, Dr. Polonsky argued that excessive fear can become a problem, particularly when it results in patients modifying their diabetes self-management for maximal hypoglycemia prevention at the expense of overall diabetes care (i.e. underdosing insulin at meals, constant snacking, swift carbohydrate consumption at the first hint of a “funny feeling,” and limiting physical activity and normal life activities). According to a Dr. Polonsky argued that a loss of confidence in one’s body is often the underlying issue that contributes to fear of hypoglycemia, especially in people with a history of anxiety. To address this loss of confidence, he discussed two broad approaches: (i) techniques that help people with diabetes to be safe by reducing the risk of hypoglycemia and (ii) techniques that help people with diabetes feel safe – namely, via behavioral approaches. On the first point, Dr. Polonsky advocated making good use of available medications and diabetes technology. While diabetes technology – namely, CGMs – can help patients identify symptoms and times of higher risk, Dr. Polonsky argued that this typically isn’t enough for people are very hypoglycemia averse. Turning to the second point, he stressed the importance of making these patients feel safe through symptom clarification, practicing treatment of hypoglycemia, and graduated behavioral exposure. For instance, he presented a case study of a 48-year-old woman living with type 1 diabetes for 10 years who began to sweat and “feel funny” when asked to describe her last episode of severe hypoglycemia. When Dr. Polonsky introduced a simple relaxation exercise, her symptoms disappeared, and she was able to recognize that she was not actually experiencing low blood glucose; rather, she was just feeling anxious. In addition to relaxation training, Dr. Polonsky also suggests that patients practice correcting by carb intake when not experiencing hypoglycemia by consuming a favorite fast-acting carbohydrate treat and noting how long it takes for their blood glucose to rise at least 50 mg/dL. Finally, graduated behavioral exposure – such as slowly introducing physical activity, reintroducing appropriate medications, and reducing nighttime snacking – can help people with fear of hypoglycemia return to normal life and social activities. Overall, Dr. Polonsky highlighted how behavioral approaches, coupled with CGM use, can help restore patients’ confidence in their ability to manage hypoglycemia episodes in order to ultimately reduce fear of hypoglycemia and improve diabetes management.

Dr. Simon Heller advocates use of diabetes technology in combination with structured education to reduce hypoglycemia incidence and improve hypoglycemia awareness

The esteemed Dr. Simon Heller (University of Sheffield) advocated for a combination approach to improving impaired awareness of hypoglycemia through the use of diabetes technology alongside structured education. Dr. Heller argued that hypoglycemia remains a key barrier to improved glycemic control, citing a study that found an average of 2.4 severe hypoglycemia events per person-year among people with type 1 and type 2 diabetes in North America. Over one-third of people with diabetes (37%) experienced at least one severe hypoglycemia event per year, with this number increasing to 51% in those with type 1 diabetes. Older people with diabetes and those with longer diabetes duration are at a higher risk of impaired hypoglycemia awareness (IAH). IAH affects 25% of those with type 1 diabetes and 10% of those with insulin-treated type 2 diabetes, resulting in six- and 17-fold higher rates of severe hypoglycemia in type 1 and type 2 diabetes, respectively. Given the high prevalence of severe hypoglycemia and especially IAH, Dr. Heller cautioned against “being too satisfied with technology and education” and advocated a combination approach to restore awareness and reduce severe hypoglycemia. On interventions, Dr. Heller briefly touched on novel insulin analogues (including inhaled insulin) and CGM, both of which have been shown to improve glycemic control and reduce the risk of hypoglycemia but do not lead to changes in IAH. For instance, in the SMILE study of hybrid closed loop, MiniMed 640G (predictive low glucose suspend) produced a remarkable 84% reduction in severe hypoglycemia in a high-risk population of people with long duration of type 1 diabetes. However, there were no significant changes in Clarke or Gold scores for hypoglycemia awareness. Meanwhile, structured education programs offer some hope for improving hypoglycemia awareness among people with IAH. In particular, Dr. Heller highlighted the DAFNE (Dose Adjustment for Normal Eating) program in the UK,  which teaches the basics of intensive insulin therapy, and led to improved awareness in those with IAH. Indeed, a 2015 meta-analysis by Yeoh et al. found that both educational and technology interventions improved glycemic control and reduced severe hypoglycemia, but only some educational programs improved awareness and most technology interventions did not improve awareness. In closing, Dr. Heller highlighted the HARPDoc study as a promising educational program to address harmful cognitions surrounding hypoglycemia, with additional mental health benefits as seen in the reduction in scores for diabetes distress, anxiety, and depression.

• Addressing the inconsistent results for interventions to improve hypoglycemia awareness, Dr. Heller emphasized that IAH is a heterogeneous condition that can arise through different underlying pathologies. Broadly, he described two main forms of IAH: (i) IAH that arises due to the underlying pathogenesis of long diabetes duration and (ii) IAH that arises due to a maladaptive stress response, in which the counterregulatory hormonal response to hypoglycemia is suppressed, due to repeated acute hypoglycemia. Given the heterogenous nature of IAH, Dr. Heller stated that it’s naïve to expect a single intervention to improve awareness and ultimately called for larger multi-clinical trials that seek to define the phenotype of IAH as a precursor before evaluating potential interventions.

Dr. Katharine Barnard-Kelly champions recognition of PROs to inform robust, rigorous, and personalized diabetes management

During an ISPAD parallel session, Dr. Katharine Barnard-Kelly (Barnard Health, UK) discussed diabetes care in the context of a biopsychosocial model of care. After highlighting the widening gap in access to diabetes technology in the UK and citing that children from disadvantaged backgrounds can be up to ten times less likely to have access to diabetes technology than non-disadvantaged peers, Dr. Barnard-Kelly noted that healthcare provider bias is a contributing factor to such disparities in technology access. We couldn’t agree more, and hope you’ll read our Diabetes UK 2022 coverage for more on how providers can act as gatekeepers to diabetes technology. She then went on to describe health as “beyond the physical,” emphasizing that mental and social wellbeing also must factor into one’s definition of health. She remarked that while diabetes technology is designed to alleviate both physical and emotional burdens for users, high expectations for such devices can negatively impact quality of life because “people want [and expect] more than optimal diabetes control.” PROs (patient reported outcomes), Dr. Barnard-Kelly argued, can account for the complex, intermingling determinants of health and allow clinicians to better support their patients while fostering intrinsic motivation that can lead to positive behavior change. She introduced two evidence-based behavior change theories: (i) self-determination theory; and (ii) social cognitive theory, which both point to self-perceptions, beliefs, and attitudes as key influencers of behavior change and motivation. Using these theories, PROs can highlight the specific feelings, emotions, and beliefs of patients, which can all be used to provide personalized support that leads to effective behavior change. Dr. Barnard-Kelly described one such PRO, the FDA-approved INSPIRE questionnaire, which identifies patient perceptions of the impact of AID systems on one’s psychosocial wellbeing and allows AID users to better manage their diabetes.

• Dr. Barnard-Kelly argued that psychosocial burdens from using diabetes technology can often be overshadowed by glycemic outcomes. Dr. Barnard-Kelly cited data from a qualitative study, in which participants (n=39 type 1s and 2s aged 10-25 years old and n=44 parents) took part in semi-structured interviews on what their ideal AID system would look like (Commissariat et al., 2021). Study participants reported wanting to minimize the physical and emotional burdens of AID system use, including lessening the user responsibility for managing glucose levels, mitigating negative emotional reactions the sound or frequency of alerts, and increasing feelings of normalcy.
• Dr. Barnard-Kelly highlighted several PROs used in pediatric diabetes as an example of how they can be effective tools. Per Dr. Barnard-Kelly, PRO measures such as PAID, PedsQL, DQOL-Youth, INSPIRE, and Spotlight-AQ can be useful for managing the transition from childhood to adulthood and offering better social support for adolescents and young adults with diabetes as they gain more responsibility and independence over their diabetes management.

Panel of women leaders in diabetes discuss unique challenges faced by women with diabetes across the lifespan; highlight potential of diabetes technology in understanding sex- and gender-based differences and individualizing treatment; call for further funding to back research

In one of the final sessions of ATTD, Dr. Eda Cengiz (UCSF), Dr. Lisa Gonder-Frederick (UVA) Dr. Chiara Fabris (UVA), Dr. Sue Brown (UVA), and Dr. Shylaja Srinivasan (UCSF) drew attention to technology’s potential in addressing distinct health needs of women with diabetes and the imperative need for funding this research – a topic we covered in our report for International Women’s Day in March. Covering a gamut of topics that affect women with type 1 and type 2 diabetes including health outcomes, psycho-behavioral barriers, and menstruation, they illustrated both how far we’ve come and how far we have to go when it comes to research, screening, and interventions that address and account for these intricacies. Of course, people with diabetes across gender identities may experience concerns related to these topics, and technology should aim to improve and personalize treatment outcomes for all of them. A major limitation is the paucity of research and existing studies’ limited sample sizes. We are excited to see continuing work led by these trailblazing researchers and many others. Powerful larger datasets such as those compiled by the Tidepool Period Project and the Beta Cell Foundation could become instrumental tools in driving understanding and technology development forward.

• Dr. Cengiz reviewed the existing evidence on the burden of diabetes and diabetes-related health complications in women, emphasizing the need for further research on the interplay of hormones, genes, lifestyle, and environment in diabetes pathophysiology. Women with type 1 and type 2 diabetes face higher risk of morbidity and mortality from cardiovascular disease than men, while men with diabetes have faster progression of microvascular complications. However, guidelines for diabetes treatment do not differentiate based on sex- or gender- differences, and high-quality research on mechanisms and outcomes is necessary to improve prevention, diagnosis, and treatment. Dr. Cengiz noted the difference between sex differences – which are biology-linked – and gender differences – which involve sociocultural, environmental, nutritional, attitudinal, and lifestyle factors. Both are implicated in different outcomes experienced by women with diabetes, although their different contributions to outcome disparities are not well-understood; this remains a key gap in research. For example, the excess vascular complication risk observed among women with diabetes compared to men could be due to both biological and/or gender-related factors. Biological factors may include hormonal and vascular pathophysiology (notably, the lost benefits of estrogen on reducing cardiovascular disease risk compared to women without diabetes), but social factors like stigma and culture may affect access to health services and health-seeking behavior. Dr. Cengiz advocated for diabetes technology as an essential means to address these disparities, through improved data collection and analysis, optimized device design, and precision medicine to address unique and specific health needs for PWDs across gender identities to improve their treatment outcomes.
• Dr. Gonder-Frederick discussed numerous psycho-behavioral challenges that are disproportionately experienced by women with diabetes. Women with type 1 diabetes face a significantly higher incidence of eating disorders, depression, anxiety, and diabetes distress compared to men, all of which may negatively impact self-care and diabetes management. For some women with diabetes, pregnancy can be a time that these challenges surface; qualitative research exploring women’s experiences of pregnancy-related diabetes management found themes of high stress levels surrounding tight glycemic target goals, guilt, challenges with eating behavior, and distress associated with the medicalization of pregnancy. Dr. Gonder-Frederick highlighted the importance of screening for these problems so they can be addressed early, improving management and long-term outcomes. On tech, she noted that it is not well-understood whether technology mitigates or exacerbates some of these challenges or how psychosocial barriers may affect adoption and usage of technology among women with diabetes. Furthermore, she made clear that in studying these issues, it is also important to note that research should go beyond gender comparisons; women are not a homogenous group, and future work should take into account intersectional identities and individual differences among women.
• Dr. Fabris and Dr. Brown from the UVA Center for Diabetes Technology presented compelling data on the impact of the menstrual cycle on insulin sensitivity and glycemia in women with type 1 diabetes and how AID could be a viable means of improving glycemic management for PWDs who menstruate. Generally, studies point to decreased insulin sensitivity and premenstrual hyperglycemia in the luteal phase and increased risk of hypoglycemia during menstruation among some premenopausal women with type 1 diabetes, although this is observed only in a subsection of this population. Overall, AID appears to mitigate this variability, but studies are under-powered, and more research is needed.
  
  • A 1992 hyperglycemic hyperinsulinemic clamp study in women with type 1 diabetes (n=16) found that the seven participants who reported premenstrual hyperglycemia experienced a decrease in glucose metabolism and significant rise in estradiol between the follicular and luteal phases, which was not observed in the nine participants who did not report premenstrual hyperglycemia.
  • In a small pilot study of women with type 1 diabetes using insulin pumps (n=5), intravenous glucose tolerance tests were performed during both phases of the menstrual cycle. Three of the five participants saw up to 24% reduction in insulin sensitivity in the luteal phase, with no difference in insulin levels detected between phases.
  • A clinical trial led by Dr. Brown recorded CGM data and insulin delivery data over the course of three complete menstrual cycles among women with type 1 diabetes (n=12) and found that hyperglycemia risk was higher in the luteal phase of the cycle, suggesting that phases of the menstrual cycle should be accounted for when planning insulin therapy for women with type 1 and that these patterns should be accounted for when refining closed loop system parameters.
  • The UVA team completed an in-silico study with the University of Virginia/Padova T1D Simulator, introducing a profile of insulin sensitivity variability across the menstrual cycle from euglycemic insulin action clamp studies to analyze whether there’s room for improvement in glycemic management for women experiencing premenstrual hyperglycemia. In the NOMINAL study, default insulin therapy parameters were used across the menstrual cycle, and in the INFORMED study, therapy parameters were modified mid-cycle to adjust for variability in insulin sensitivity. In simulation results from NOMINAL, mean Time in Range dropped from 62% in the follicular phase to 40% in the luteal phase in the open loop group and from 72% to 63% in the closed loop group. In the INFORMED study, when therapy parameters were adjusted, Time in Range stabilized across the cycle for both open-loop and closed-loop groups. Clinical validation is needed to back the results of the simulation, but findings were highly suggestive that current insulin therapy is not optimized to handle variability across the menstrual cycle.

• • Recently-published data from the AID Menses Trial (secondary analysis from the Control-IQ pivotal extension study) analyzed glycemic metrics and insulin delivery throughout the menstrual cycle in women with type 1 diabetes using a closed loop system (n=15) and found that metrics remained stable across cycle phases. CGM data revealed no obvious changes in Time in Range or time below range in the luteal phase versus the menstrual phase and mean daily doses for basal, bolus, and total insulin did not change notably throughout the cycle. While this data suggests that changes in glycemia across the cycle are not evident in women who use AID, Dr. Brown emphasized that the main takeaway here is the difficulty of drawing conclusions due to small sample size and dearth of studies in this area.
  • In pooled studies of three AID trials using the Control-IQ algorithm (DCLP1, DCLP3, and Project Nightlight), there was a significant interaction between gender and treatment for Time in Range and time spent in hyperglycemia, but this did not differ at baseline or after AID use. Prior data suggests that this may be related to time spent using the AID algorithm. No significant interaction between gender and treatment for A1c was observed.
  • Secondary analysis of data from a study on one-year real-world use of Control-IQ found a significant interaction between gender and treatment (AID use) for Time in Range and time above range, and a significant interaction between gender and treatment for GMI, but Dr. Brown noted that these relationships are not likely to be clinically significant.
• Dr. Srinivasan closed the session with a focus on the needs and experiences of women with type 2 diabetes, with a call for further assessment of the impact of sex and gender on type 2 diabetes interventions. Dr. Srinivasan Built on Dr. Cengiz’s presentation, stating that technology has been vastly understudied as a means of addressing the lifelong impact and interaction between sex and gender on type 2 diabetes risk and disease burden. Studies of CGM, smart pen, and pump technologies in type 2 populations show glycemic and quality-of-life benefits but have not reported sex-specific effects or interactions. Some data on type 1 and type 2 pregnancies suggests that CGM use reduces risk of adverse neonatal outcomes in women with type 2, but there is a lack of studies that specifically look at CGM use in pregnant women with type 2 diabetes.

Dr. Laurel Messer discusses daily predictors of diabetes self-management and goal-setting in adolescents and young adults with type 1 diabetes

University of Colorado’s highly regarded Dr. Laurel Messer presented results from a questionnaire-based study assessing glycemic control and diabetes self-management in adolescents and young adults with type 1 diabetes. A staunch advocate of comprehensive education to support technology use among youth with type 1 diabetes (especially for AID), Dr. Messer emphasized that all diabetes technologies rely on diabetes self-management behaviors. Dr. Messer noted that diabetes self-management may not necessarily be at the top of adolescents’ minds, eloquently stating that “it’s difficult to attend to diabetes management when you’re trying to figure out who you are as a person,” on top of balancing school, family, and other responsibilities. Although aggregate data can reveal broad trends in glycemic control, Dr. Messer argued that such graphs hide the individual day-to-day variation that is characteristic of adolescence.

To address questions about how diabetes care changes when we’re stressed, depressed, or not sleeping well and why patterns of glycemic control are different each day, Dr. Messer designed a study to measure daily variation in glycemic control, diabetes self-management, and goal attainment and to evaluate whether daily factors could predict these three variables. Dr. Messer’s Daily Predictors of Diabetes Self-Management study examined glycemic control, diabetes self-management, and goal attainment in 88 adolescents and young adults (ages 14 to 26 years) with type 1 diabetes who were on CGM. The two central aims were to: (i) characterize daily variability in glycemic control, diabetes self-management, and goal attainment and (ii) prospectively identify daily factors that predict the above variables based on data from the start of the day. At the start of the study, participants were asked to set a diabetes management goal for the next two weeks, such as bolusing for every meal. The two-week trial was designed with six quasi-random days in which morning engagement surveys and evening goal attainment surveys were administered. Participants had an average age of 17.6 years, average diabetes duration of 8.5 years, and an average A1c of 7.9%; 54% were female and 90% were white.

• Results showed significant variation in daily glycemic control and bolusing. Mean Time In Range fluctuated 16% and mean sensor glucose fluctuated 30 mg/dL, while the average number of daily boluses fluctuated by two per day. To build the prediction model, Dr. Messer identified seven key factors from the questionnaire: (i) current glucose level upon waking; (ii) planning to manage diabetes that day; (iii) desire to manage diabetes that day; (iv) wanting to skip diabetes self-management that day; (v) feeling good about one’s identity; (vi) rating one’s health as okay; and (vii) and needing extra support for diabetes self-management that day. The seven-item model was able to predict 17% of Time In Range and 18% of mean glucose for a particular day. Meanwhile, the model only predicted 2% of the variance in day-to-day variability in number of boluses administered. Prediction of hyperglycemia response was slightly better, with the seven-item model explaining 14% of variability here. The model was most successful in goal attainment, predicting 29% of variability in whether participants met their diabetes self-management goals each day. Ultimately, Dr. Messer advocated for greater focus on the daily experience of people with diabetes, particularly adolescents and young adults with diabetes, in order to identify opportunities for successful behavioral interventions.

Point-of-Care A1c testing and diabetes diagnosis

Dr. David Sacks (NIH Chief, Clinical Chemistry Service) argued that point-of-care (POC) A1c is not adequate to diagnose diabetes. Citing a host of studies (see slide below), he explained that POC A1c devices are not accurate or precise and that there is no evidence to support the clinical performance of POC A1c devices for the diagnosis of diabetes. He cited a 2017 meta-analysis to show that POC A1c devices are both inaccurate and imprecise for the diagnoses of diabetes and that there are significant differences between A1c results by POC devices and lab instruments. He added that POC A1c devices are unable to detect the presence of hemoglobin variants, unlike lab tests. Furthermore, he pointed out that POC A1c devices have minimal regulatory oversight with no regular inspections, whereas labs are inspected every two years. POC A1c devices fall in a category of testing in the US called “waived tests,” which means they do not require proficiency testing or regular assessment of accuracy or precision. Dr. Sacks cited a 2001 CMS report and a 2005 CDC report showing that locations performed waived tests did not follow the proper procedures to ensure accurate results. 

Personalization, structure, engagement, and social support identified as key factors in behavioral weight loss interventions

Ms. Anne Wolf (University of Virginia) discussed key components of behavioral therapy for weight loss and clinical pearls to maximize effectiveness. Ms. Wolf specifically explored the role of technology in improving patient outcomes, as the challenges that often prevent weight loss program participants from reaching their goals are unique. Ms. Wolf advocated for technology use to reduce cost, connect coaches with participants, and improve patient engagement in order to increase retention for patients while reducing the burden on providers. Ms. Wolf reviewed evidenced from the National Diabetes Prevention Program, which as many know, included 22 group sessions following an evidence-based curriculum over a year long period, ultimately producing average weight loss of 4% (just under the generally accepted 5% threshold of clinical efficacy). While subsequent interventions have had varying degrees of success, Ms. Wolf identified personalization, structure, frequent engagement, and social support as key features that tended to have more positive weight loss results. Additionally, Ms. Wolf said that patients tend to have greater participation and higher retention if they see immediate results, arguing for support for participants early in the process to keep them excited about the program. Generally, though, the weight loss conferred by behavioral interventions does not exceed 5%. Thus, we continue to wonder which patients are an appropriate candidate for these intensive programs and which patients might benefit more from a new weight loss medication that can induce much greater weight loss (ex. 15% for semaglutide, 22.5% for tirzepatide!) This is a question we will continue to explore as we learn more about as the obesity treatment paradigm adjusts to reflect these new pharmacological options.

It doesn't get better than the ATTD 2022 Tech Fair: Updates from AMF Medical, Capillary Biomedical, PharmaSens, Undermyfork, and PVRmed

ATTD’s Tech Fair highlighted some of the most exciting smaller companies in and around diabetes technology. While these companies are smaller, some are certainly familiar names, and we enjoy these periodic updates on their progress! Sitting at the Tech Fair in the middle of ATTD’s buzzing Exhibit Hall, we couldn’t help but think, “Innovation is everywhere – just look around!” Also, make sure to check out our highlight on Diabeloop’s presentation at the Tech Fair, which we broke out separately.

AMF Medical

The first company in the ring was AMF Medical, a company that has been busy around the recent conference circuit revealing its Sigi insulin patch pump. This company first came on our radar in February 2021 and was also one of our highlights at last year’s ATTD 2021 Tech Fair. While last year’s Tech Fair presentation focused on the technology behind Sigi’s highly accurate insulin delivery and fast occlusion detection, this year, presenter Dr. Antoine Barraud (co-founder and CEO) highlighted how that technology, coupled with the pump’s innovative design, creates a unique user experience that differentiates Sigi from other pumps on the market. In particular, Dr. Barraud identified the three biggest burdens with currently available insulin pumps: time, thinking, and space. On the time front, Dr. Barraud showed Sigi’s “one-click, one-hand applicator” and its use of a pre-filled insulin cartridge that is “plug-and-play” (no waiting for insulin to reach ambient temperature). To reduce thinking burden, Dr. Barraud noted that Sigi is designed to be used with AID systems in the future. Finally, on the space burden, Sigi is smaller and thinner than Insulet’s Omnipod and Roche’s Accu-Chek Solo pumps. AMF intends to offer Sigi pump users full smartphone control, which would eliminate the need for a secondary controller. We were particularly impressed by one image that demonstrated the engineering challenge of placing a pre-filled insulin cartridge in such a small pump – wow!

In addition to its Tech Fair presentation, AMF Medical also presented a poster at this year’s ATTD. The study results demonstrated Sigi’s ability to dose controlled amounts of basal and bolus insulin with very strong accuracy. Additionally, Sigi’s occlusion detection time was 29 minutes with a basal rate of 0.1 U/hr and just 10 minutes with a basal rate of 1.0 U/hr.

Capillary Biomedical

At Capillary Biomedical’s tech fair presentation, CEO Mr. Paul Strasma updated attendees on the company’s SteadiSet insulin infusion set. Mr. Strasma highlighted some of the differentiating features of Capillary Bio’s device, primarily focusing on the company’s SteadiFlow cannula. The SteadiFlow cannula is made from a softer material than traditional cannulas and includes an interior coil that is designed to reinforce the cannula and prevent kinking. Additionally, the cannula has three side holes allowing insulin delivery to continue even when the tip of the cannula is occluded. Studies in pigs have also suggested that SteadiFlow’s angled insertion significantly reduces the amount of inflammation compared to a traditional 90-degree insertion cannula. This reduced inflammation could be particularly important for maintaining consistent insulin delivery and absorption for longer periods of wear (e.g., seven days). As a reminder, the company presented a feasibility study (n=41 infusion sets) at ADA 2021 showing a very strong survival rate of 88% out to seven days. Capillary Bio, which is now up to 25 employees, has previously received a $1.5 million loan from the Helmsley Charitable Trust for feasibility work and is now looking to run a pivotal trial of its SteadiSet device, with FDA clearance projected for 1H23.

PharmaSens

PharmaSens’s presentation centered on the company’s niia System Concept, the center of which is the niia Patch System. Marketed as “one patch for all needs,” the patch has a combined insulin pump and a CGM. From the company’s website, we learned that this combined CGM and pump is the company’s “niia Signature” device. It also appears that the company has a “niia essential” basal-bolus patch pump that requires no smartphone control (our minds went to CeQur), and also the niia advanced standalone pump with remote smartphone control. We got a look at the company’s pipeline (see picture below), which shows that the company also has a fourth device (“simplex”) in the works. PharmaSens is currently preparing to submit its niia essential System for CE-Marking in 2023, with an eventual launch targeted for 2024.

Undermyfork

Undermyfork Co-Founder and COO Mr. Mike Ushakov highlighted the “Undermyfork Care” provider platform, which was announced at DTM 2020 and launched in February 2021. As a reminder, the Undermyfork app pairs CGM data with automated food-logging data to help users identify foods that help keep them in range and foods that tend to drive them out of range. The app can semi-automatically tag foods by suggesting tags to users based on image recognition (e.g., brown rice, yogurt, toast, etc.). Retrospectively, patients’ meals are automatically color-coded based on the two-hour postprandial Time in Range: green for 75%-100%, amber for 40%-75%, and red for 0%-40%. Undermyfork Care is the company’s provider-facing web-based platform that launched in February 2021 and allows providers to view their patients’ meal, insulin, and CGM data all in one place. As with the app, providers can filter a patient’s data by time period, by postprandial Time in Range (green, amber, and red), and by meal tags. Providers can also view patient data as a daily or monthly report that shows their CGM data, as well any added meals or insulin doses. During his presentation, Mr. Ushakov described the Undermyfork app (iOS; Android) and the Undermyfork Care provider platform, specifically stressing that the latter can give providers a “helicopter view” of their patients, and that the platform’s PDF reports can easily be incorporated into EHRs. We loved seeing Mr. Ushakov’s live demo, during which we saw how a user who eats the same meal every Monday can compare their CGM traces across different instances of that same meal. Undermyfork is an A+ app from our view – Kelly’s been using it for some time, loves it, and we first heard about it from Close Concerns alum Adam Brown back in 2019.

PVRmed

PVRmed is a company tackling the ever-so-relevant topic of diabetic foot ulcers (DFUs) and peripheral arterial disease (PAD). With a mission to make “the world free of diabetic foot complications,” the company’s system stimulates neo-angiogenesis by transporting carbon dioxide noninvasively into one’s lower extremities, which induces local hypoxia that ultimately leads to enhanced peripheral circulation via microcapillary formation. The device is quite big (see the picture below). The company pointed to four peer-reviewed studies in support of its device: (i) Macura et al., 2020; (ii) Finzgar et al., 2015; (iii) Frangež et al., 2017; and (iv) Frangež et al., 2021. We loved seeing the picture the company provided showing ulcer healing over time using the company’s device – it’s simply miraculous what it has been able to achieve! Lower-limb-related issues comprise 33% of US diabetes expenses, and each amputation can cost up to $100,000, meaning that preventative care could lead to massive savings for the entire healthcare ecosystem. We salute PVRmed and all that the company is aiming to do for people with diabetes.

Poster Tables

Closing the Loop and Automated Insulin Delivery

Title	Details + Takeaways
Improvement In HbA1c after 8 Weeks of Omnipod 5 Automated Insulin Delivery System Use in Adults with Type 2 Diabetes: From Injections to Closed-Loop Therapy	Omnipod 5 type 2 feasibility study; on MDI (n=12) or basal-only insulin therapy (n=12) at baseline; used OP5 for eight weeks Among those on MDI at baseline, Time in Range +3.4 hours/day to 61% and A1c –1.2% to 8.1% Among those on basal-only insulin at baseline, Time in Range + 5.9 hours/day to 57% and A1c -1.4% to 8.1% Massive declines in time in severe hyperglycemia (>250 mg/dL) in both groups (-2 hours/day to 9.3% in MDI group; -5 hours/day to 12% in basal-only group
User Experience Of The Omnipod® 5 Automated Insulin Delivery System In Adults With Type 2 Diabetes	Human factors data from 14 participants in Omnipod 5 type 2 feasibility study; participated in 1:1 semi-structured interviews Reported system usability score of 90.5 out of 100 Reported significant increase in satisfaction with their insulin delivery (insulin delivery satisfaction score improved from 3.7 to 4.4)
Glycemic Outcomes and Safety with MiniMed 780G System in Children with Type 1 Diabetes Aged 2-6 Years	Open-label, prospective study (n=35) of preschoolers (ages 2-6) on MiniMed 780G for 12 weeks; were compared to 35 matched historical controls Average age of four; mean baseline A1c of 7.5% (8.7% among those on MDI, 6.9% if on 640G, 8.7% if on 670G); most on 640G (54%) or 670G (37%) at baseline Time in Range +2.1 hours/day from ~58% to ~67% A1c, mean glucose, and time above range also improved during follow-up No change in time <70 mg/dL (3% vs. 3.2%; p=0.51) or glycemic variability (CV of 38% in both; p=0.33) No DKA or severe hypoglycemia
Advanced Hybrid Closed-Loop Study in an Adult Population with Type 1 Diabetes (ADAPT): A Randomized Controlled Study	Six-month, multicenter RCT comparing MiniMed 780G (n=41) vs. MDI + FreeStyle Libre (n=41) in adults with type 1 with baseline A1c ≥8% 780G users’ A1c fell from 9% to 7.3% at three months and was maintained at six months, whereas no significant change in MDI group; at six months, 780G users saw -1.4% A1c improvement relative to MDI/FSL group (p<0.0001) At six months, 780G users achieved TIR of 71% vs. 43% with MDI/FSL (+6.7 hour/day TIR with 780G vs. MDI) Far more 780G users achieved targets than MDI/FSL users: 28% vs. 0% for A1c <7%; 58% vs. 7% for GMI <7% No safety concerns, satisfaction scores higher in 780G group
Transition of Patients with T1D From MDI And SMBG Directly To MiniMed 780G Advanced Hybrid Closed Loop System: Results of a Two-Center, Randomized Controlled Study	Three-month RCT evaluating MiniMed 780G (n=20) vs. MDI/BGM (n=17) in those naïve to CGM and pump tech (baseline A1c 7.2%, age 40) Time in Range increased from 69% to 85% with 780G, no change in control group (+5.2 hour/day relative improvement) Time <70 mg/dL significantly improved with 780G (8.7% to 2.1%) vs. no change in control group (7.5% to 8.1%); treatment benefit of -1.1 hour/day -0.6% A1c improvement with 780G relative to control All other CGM outcomes favored 780G
A Comparative Study Using Control-IQ Closed Loop System vs Basal-IQ System in Very Young Children with Type 1 Diabetes (T1D): Clinical Effectiveness And Safety	Prospective study comparing Control-IQ vs. Basal-IQ in preschoolers with type 1 (n=10, average age of 4.6) Time in Range improved +3.4 hours/day with Control-IQ vs. Basal-IQ (77% vs. 63%; p<0.05); time >250 mg/dL improved -2.5 hours/day (2.6% vs. 13%; p=0.01) Glycemic variability improved with Control-IQ vs. Basal-IQ (32% vs. 41%; p=0.002) No severe hypoglycemia or DKA
Evaluation of Hybrid Closed-Loop Insulin Delivery in Patients with Type 1 Diabetes in Real-Life Conditions: Previously Used Therapy Matters?	One-year, multicenter observational study evaluating outcomes on MiniMed 670G across various different baseline insulin delivery systems in type 1s (n=60) in Argentina Baseline insulin delivery method (18% 640G, 10% MiniMed 754, 33% MDI, 15% Accu-Chek Combo, 21% MiniMed Veo, 1.6% MiniMed 715) did not impact glycemic improvements at 12 months Across all participants, A1c decreased from 7.8% at baseline to 6.7% at three months (p=0.02), maintained at 7.1% at 12 months (p=0.02) Time in Range improved from 57% at baseline to 72% at one month, maintained >70% out to 12 months (p=0.01) Time <70 mg/dL fell from 5% at baseline to 2.6% at three months, maintained <3% throughout 12 months of 670G use (p=0.01) in all groups
Long-Term Outcomes on Glucose Control, Sleep, and Health Economy after Implementation of Tandem Control-IQ in a Pediatric Population	12-month, observational controlled study evaluating Control-IQ vs. MDI/Dexcom G6 vs. pump/G6 in children (ages 6-18, n=84) in Sweden Time in Range was significantly improved with Control-IQ vs. other arms during all months after start Those on Control-IQ on average achieved TIR/TAR/TBR targets across entire year of use Analyses ongoing to evaluate impact on sleep, sick leave in parents, health economics
Glycemic Outcomes in Pediatric and Adult Individuals with Type 1 Diabetes (T1D) during MiniMed 780G System Use with the Guardian 4 Sensor	Three-month extension phase of MiniMed 780G pivotal with Guardian 4 in 176 type 1s (ages 7-75) 95% time in Auto Mode Achieved A1c of 7.1% at baseline (7.2% for ages 7-17; 6.8% for ages 18-75); Time in Range of 73% (72% for ages 7-17; 77% for ages 18-75) Active Insulin Time of 2-3 hours led to greatest TIR while also maintaining recommended time <70 mg/dL of <4% Results showed that improvements seen in pivotal w/Guardian Sensor 3 were maintained with Guardian 4
Calibration, Metabolic Control and Sleep Quality in MiniMed 780G Type 1 Diabetes Cohort with Guardian Sensor 3 and 4	Cross-sectional observational study of MiniMed 780G users (n=37; 15 pediatrics, 22 adults) when switching from Guardian Sensor 3 vs. Guardian 4 No significant differences in glycemic outcomes with Guardian Sensor 3 vs. Guardian 4 Significantly fewer BGM readings/day with Guardian 4 vs. GS3 (1.7 vs. 3.6); significant fewer calibrations/day (1.4 vs. 3.3)
Differences in Glycemic Control between Distinct Closed-Loop Systems in Real Life	Longitudinal, real-world comparison of MiniMed 780G (n=18) vs. Control-IQ (n=18) vs. Diabeloop (n=54) over three months Not randomized so there were significant differences in the groups’ baseline characteristics in terms of age, baseline use of insulin bolus advisor, and educational level, which may have been confounders in 780G group’s advantage Glycemic improvements seen across all treatments but greatest GMI improvement with MiniMed 780G, plus trend toward relative improvement in TIR, TAR, TBR No between-group significant differences in A1c at three months
Reasons for Hesitancy toward New Automated Insulin Delivery System Adoption among Adults Living with Diabetes in the United States, Canada, and Europe	Survey of 5,226 adults with diabetes in US, Canada, France, Germany, Italy, Netherlands, Sweden, UK taken in March-June 2021; evaluated qualitative responses from 808 participants who were not interested in Control-IQ, MiniMed 780G, or Omnipod 5 and offered explanation Top reasons for AID system reluctance: unwillingness to switch therapies (19%), unwillingness to use insulin pump (16%), distrust of technology (12%) DIY AID system users found the settings too restrictive (57%) Among those on MDI, 17% did not want to use pump and 20% preferred injections US participants more distrustful of technology (15%) than Canadian and European participants (7%; p<0.001)
Performance of the Omnipod 5 Automated Insulin Delivery System with and without Pre-Meal Bolus	Three-month, multicenter, single-arm study evaluating Omnipod 5 with and without pre-meal boluses in children (ages 7-13; n=110) and adolescents/adults (ages 14-70; n=80); participants were their own comparator group using same meal at same time on different days (one with meal bolus, one without meal bolus) Without meal boluses, children and adults saw substantial increases in mean glucose level Without meal boluses, children increased from 129 mg/dL to 241 mg/dL 1.8 hours post-meal, didn’t fall <180 mg/dL until 3.6hr post-meal  Without meal boluses, adolescents/adults saw glucose rise from 125 mg/dL to 214 mg/dL 1.7 hours post-meal, didn’t fall <180 mg/dL until 3.4 hours post-meal
Time of Initiation of Advanced Hybrid-Closed Loop Therapy and Related Glycemic Outcomes in People with Type 1 Diabetes Transitioning from Multiple Daily Injections	Retrospective study evaluating impact of time of initiation of Control-IQ following pump start on outcomes for those transitioning from MDI (n=17,540) Three groups: group 1 (n=14,222) started Control-IQ w/in 2 days of start pump; group 2 (n=2,448) started Control-IQ 2-14 days after starting pump; group three (n=870) started Control-IQ 15-90 days after pump start; no significant baseline A1c differences between groups (8.5% in all) No between-group difference in time <70 mg/dL in first 90 days of use (1 = 0.87%; 2 = 0.84%; 3 = 0.89%); same was true of time <54 mg/dL No significant difference in Time in Range between group 1 and 2 (69% vs. 70%), but both significantly higher than group 3 (67%; p=0.003), suggesting that those who initiated Control-IQ within first 14 days of pump use saw best results
Changes in Quality of Life and Psychological Well-Being in Patients with T1D Undergoing Transition from MDI And SMBG Treatment Directly to MiniMed 780G ACHL System	Three-month RCT evaluating QoL impact of transitioning from MDI/BGM to MiniMed 780G vs. staying on MDI/BGM in 41 type 1s 780G group reported improvements in feeling well, working, eating as I would like, and doing normal things, as well as reduced anxiety levels
Who is Using Do-It-Yourself Artificial Pancreas Systems and Why	Survey of 662 people with type 1 using DIY in US (44%), Canada (15%), and UK (11%) recruited via social media 87% White; 87% post-secondary education; 70% using system themselves, 30% caregivers to children using system 70% using Loop, 23% on AndroidAPS, 2% on OpenAPS, remainder not reported When asked why use a DIY system rather than commercial system, reasons included: transparency (33%), interoperability (29%), support for open-source software (25%), only AID available where live (22%), belief in greater safety (19%), belief that is more affordable (12%)
Which Do-It-Yourself Artificial Pancreas Systems (DIYAPS) And Used in the United Kingdom? Insights from the Association of British Clinical Diabetologist’s (ABCD) Audit Programme	ABCD Audit of DIY users in UK; poster includes results from first 101 audited users 90% were White British, mean diabetes duration 26 years, age 41, A1c 7.5% Majority (61%) on AndroidAPS, 27% on Loop, 12% on OpenAPS 81% of pumps used were NHS-funded; most used includes DANA (40%) followed by Omnipod (18%), Roche Combo (14%), Roche Insight (13%) Majority (59%) of CGMs were NHS-funded; FreeStyle Libre + MiaoMiao most used (43%) to convert FSL to rtCGM, followed by Dexcom G6 (39%)
What Do Preschool Children with Type 1 Diabetes (T1d) Gain by Using Advanced Hybrid Closed Loop System?	Prospective six-week study evaluating transition from SAP with predictive low glucose suspend to MiniMed 780G in 11 preschoolers (ages <7, average age 5.7 years) Time in Auto Mode increased from 87% with SAP+PLGS to 96% GMI fell from 6.8% 6.5% within first two weeks of MiniMed 780G use, maintained out to four weeks Time in Range increased from 46% to 54% (+1.9 hours/day) Trend toward a slight increase in time <70 mg/dL (4.3% at baseline to 6% at two weeks to 5.5% at four weeks)
Remote and in Clinic Initiation of Advanced Hybrid Closed Loop System MiniMed 780G in Children and Adolescents with Type 1 Diabetes	Three-month non-randomized prospective comparison of in-person (n=36) vs. remote (n=28) initiation of MiniMed 780G in children and adolescents (ages 7-18) in Qatar No significant differences in A1c, Time in Range, or Auto Mode use post-780G initiation between remote vs. in-person initiation A1c fell from 8.2% to 6.2% in in-clinic group; 8.3% to 6.7% in remote group; significant improvements from baseline but no between-group difference Time in Range improved +7.2 hours/day in in-person group (48% to 78%) and +6 hours/day in remote group (49% to 74%); significant improvements from baseline but no between-group difference
Improved Glycemic Outcomes after 6 Months of Use in Real-World of an Advanced Hybrid Closed-Loop System in Adolescents and Adults with Type 1 Diabetes	Six-month prospective study evaluating transition from MiniMed 640G + FreeStyle Libre 2 to MiniMed 780G in 47 adolescents and adults (mean age 41; 4 adolescents; baseline A1c 6.9%) Used glucose target of 100 mg/dL and active insulin time of three hours Time in Range improved + hours/day from 65% at baseline to 74% at six months (p=0.001) Time <70 mg/dL fell – hours/day from 4.6% to 2.3% (p=0.002)
Real-World Use of Control-IQ Hybrid Closed-Loop System in Type 1 Diabetes	Retrospective analysis of the transition from Basal-IQ to Control-IQ in 34 type 1s (mean age 46, 24 years of diabetes, 71% female) Time in Range increased 2.8 hours/day from 69% on Basal-IQ to 81% on Control-IQ (p<0.001) No significant change in time <70 mg/dL or <54 mg/dL, although were already meeting consensus targets at baseline GMI fell 0.3% from 7% to 6.7% (p<0.001) Far more achieved TIR ≥70% when transitioned to Control-IQ (88% vs. only 36% on Basal-IQ)
Safety and Glycemic Control in Chinese Adolescents and Adults with Type 1 Diabetes (T1D) in a MiniMed 770G System Clinical Trial	Four-week prospective evaluation of the transition from SAP to MiniMed 770G in Chinese type 1s (50 adults, 12 adolescents; baseline A1c 7.1%) Time in Range improved +1.4 hours/day from 75% to 81% (p<0.001) Time <70 mg/dL fell 36 minutes/day from 4.7% to 2.2% (p<0.001); time <54 mg/dL fell 7 minutes/day from 0.9% to 0.4% (p<0.001) Percentage of participants achieving TBR <4% increased from 52% to 89%; percentage achieving TIR ≥70% increased from 74% to 95% Time in Auto Mode averaged 94%
MiniMed 780G Hybrid Closed-Loop System Rapidly Improves Metabolic Control in Type 1 Diabetes Individuals Previously Treated with Insulin Pump	Six-month prospective study evaluating transition from pump/BGM to MiniMed 780G in X type 1s (50 years old, 25 years of type 1 diabetes) Time in Range improved 2.6 hours/day from 64% at baseline to 75% at one month, which was maintained out to six months A1c increased significantly from 6.7% at baseline to 7% at six months Significant declines in time <70 mg/dL (2% to 1%) at six months and in time <54 mg/dL
Self-Reported Sleep Outcomes Reveal Potential Influence of Automated Insulin Delivery Systems on Sleep Disturbance Among U.S. Adults with Diabetes	Survey of 1,945 adults on intensive insulin therapy analyzed to evaluate if there exists an association between insulin delivery method and sleep quality Participants average age 54 years, 93% White, 72% Bachelor’s degree or higher; 25% on MDI/CGM, 25% on SAP, 50% on AID PWD on MDI/CGM were more likely to report mild/moderate sleep disturbance and less likely to have none to slight sleep disturbance compared to AID; also more likely to report restless sleep (31% vs. 23%), trouble staying asleep (32% vs. 25%), poor sleep quality (24% vs. 19%) PWD on SAP were more likely to report mild sleep disturbance than those on AID
Impact Of DBLG1 System on T1D Patients who Spend More than 5% of Time in Hypoglycemia in Open Loop	Retrospective evaluation of DBGL1 AID system (with Roche pump + Dexcom G6) in 45 adults with type 1 who spend >5% of time <70 mg/dL on open loop Time <70 mg/dL decreased 1.2 hours/day from 8% to 3% (p<0.001) Time <54 mg/dL fell 17 minutes/day from 2% to 0.8% (p<0.001) Time in Range increased 1.2 hours/day from 63% to 68% (p<0.001) With DBGL1, 29% of participants met consensus targets for TIR, time <70 mg/dL, and time <54 mg/dL
Results Of DBLG1 System Launch in Germany Between April and September 2021	Real-world analysis of 1,917 people who initiated DBLG1 (w/Dexcom G6 and Roche Accu-Chek Insight) in Germany in September-December 2021; average number of days of analysis was 94 days (~three months) When on DBLG1, users saw Time in Range of 73%, time <70 mg/dL of 0.9%, time <54 mg/dL of 0.1%, and a GMI of 7%; spent 95% of time in Auto Mode 60% of DBLG1 users active >70% TIR, 98% <5% time <70 mg/dL, 48% a GMI <7%; 47% achieved all three criteria
The Danish Loop-DIY Study: The Effect of Loop-DIY in Danish Children With Type 1 Diabetes Mellitus	Cross-sectional retrospective study evaluating glycemic and QoL impact of DIY Loop in eight Danish children and adolescents with type 1 and their parents A1c fell a significant -0.4% from 6.8% at baseline to 6.4% at six months after starting Loop Loop use associated with improvements in parents’ quality of sleep and in adolescent users’ self-efficacy
Durable High Glucometric Performance of The MiniMed 780G Advanced Hybrid Closed Loop System in Real World Evaluation in a Value-Based Diabetes Center (Diabeter Netherlands)	Retrospective analysis of the first six months of MiniMed 780G use in type 1s in the Netherlands (n=111; 49% age <15); no baseline data shared By month one, 780G users were achieving 77% Time in Range, maintained >75% TIR through month six Those using optimal settings (Active Insulin Time of two hours, 100 mg/dL glucose target) achieved a TIR of 80% at one month and maintained a TIR >77% out to month six without increasing time <70 mg/dL or time <54 mg/dL relative to total cohort No difference seen among those ages <15 vs. >15
Current and Future Importance of AID-Systems For People With Diabetes	Survey of 2,417 people with diabetes in Germany on attitudes and assessment of AID systems; 58% type 1s, 21% type 2s, 19% parents of children with diabetes 63% of participants rated AID as currently important (increase from 2019 report), but more believed would be important in five years (88%); more parents viewed AID as important than people with type 1; lowest among type 2s Nearly 70% felt would be more independent with AID; higher diabetes education efforts were seen by nearly 50% Respondents believed that in 10 years, 50% of type 1s will be using AID and in 15 years will reach 90% 91% of participants not using an AID system would prefer a commercial system over DIY Among DIY users, 53% of parents of children with diabetes would switch to commercial AID system vs. only 26% of type 1s
Glycemic Control after Starting Advanced Hybrid Closed-Loop System in Children with Type 1 Diabetes Mellitus	Retrospective study of MiniMed 780G initiation in 32 children with type 1 Baseline insulin delivery: 72% SAP with predictive low glucose suspend; 16% 670G; 12% MDI TIR improved + hours/day from 58% at baseline to 70% at one month (p<0.05), an improvement that was maintained out to month six GMI improved from 7.3% at baseline to 7% at month one, a significant improvement maintained out to month six Significant improvements in time >180 mg/dL and >250 mg/dL; no significant change in time 54-70 mg/dL or <54 mg/dL, although already low at baseline No adverse events
Glycemic Control in a Cohort of T1DM Pediatric Patients: Which is the Best Treatment Strategy?	Retrospective study evaluating glycemic control in 248 pediatric type 1s (median age 12) according to mode of insulin delivery Modes of insulin delivery: 70% MDI, 30% pumps (15% integrated in AID system); 80% on is-CGM vs. 20% on rt-CGM AID associated with significantly more TIR, less time >250 mg/dL, and lower glycemic variability compared to other modes of insulin delivery Nearly 60% of those on AID systems achieved a TIR >70% vs. 32% of those on SAP 65% of those on AID achieved A1c <7% vs. 51% on SAP vs. 41% on MDI
Human Factors of a Fully Implantable Bionic Invisible Pancreas: Perceived Potential Benefits and Barriers of People with Type 1 Diabetes	Human factors testing on barriers and benefits for EU project FORGETDIABETES aiming to develop a fully implantable AID system Included those on MDI/CGM (n=12), CGM/pump (n=14), and AID systems (n=12) – broke these into three groups Across all groups, the perceived benefits outweighed expected barriers; no between-group differences based on current treatment 76% said they would use the system primarily due to less treatment effort, prevention of complications, facilitation of sports; barriers included fears of complications during surgery or infection, perceived loss of control over treatment, fear of technical problems
Advanced Hybrid Closed Loop System MiniMed 780G in Children and Adolescents with Type 1 Diabetes: Does Previous Pump Experience Impact the Glycemic Control?	Retrospective analysis of transition from MDI (n=53; baseline A1c 8.6%) vs. pump (n=45; baseline A1c 7.7%) to MiniMed 780G in children and adolescents with type 1 (ages 7-18) in Qatar A1c fell from 8.6% to 6.7% in the MDI group and from 7.7% to 6.6% in the pump group at three months TIR increased +6.2 hours/day from 50% to 76% in MDI group; +4.1 hours/day from 60% to 77% in pump group 94% of Time in Auto Mode with MDI group; 96% in pump group (between-group difference not significant)
Feasibility of a Clinical Tool to Support Clinicians in Caring for Persons with Diabetes Using Automated Insulin Delivery	Feasibility assessment of Barbara Davis Center’s Panther Program Control-IQ Clinical Tool for supporting clinicians that are caring for those using Control-IQ Clinicians completed system useability scale (SUS) to assess satisfaction and usability Included 28 clinicians (MDs, NP/PA, educators), 89 clinician encounters SUS scores in 95th percentile for usability (“excellent”) Clinician confidence in providing care increased from 7.7 to 9.0 after using tool 75% reported overall satisfaction; 90% would recommend to colleagues
Sleep And Fear Of Hypoglycemia In Parents Of Youth With Type 1 Diabetes Following Implementation Of Hybrid Closed-Loop Technology	Prospective assessment of the association between parent fear of hypoglycemia and sleep measures and how these change with initiation of Control-IQ (at three and six months) Included 39 youth (age 11, 97% previous CGM use) and 39 parents (82% female) No significant associations between sleep and fear of hypoglycemia total and subscales Parent fear of hypoglycemia improved after three months of Control-IQ use and was sustained out to six months
A Short-Term Evaluation of Children with Diabetes Using Medtronic 780G System- A Single Center Experience from Turkey	Retrospective evaluation of: (i) children (n=25) who transitioned from MiniMed 640G to MiniMed 780G; and (ii) children (n=33) who initiated 780G regardless of baseline mode of insulin delivery Those who transitioned from 640G to 780G saw TIR +57 min/day from 76% to 80%; no significant change in TBR Using cohort 2, analysis showed that as bolus rate increased, TIR increased; basal rate and autocorrection rate negatively associated with TIR
Glycemic Outcomes With Open-Source Automated Insulin Delivery Systems After 12 Months Of Use In Adults With Type 1 Diabetes	Retrospective 12-month analysis of adults (n=23) using AndroidAPS (n=22) or Loop (n=1) Age 38, baseline A1c 6.6%, 79% previously used pump Time in Range improved from 70% at baseline to 88% at 12 months (p<0.001) Time >250 mg/dL decreased from 5.1% at baseline to 0.6% at 12 months (p<0.001) Time <70 mg/dL decreased from 5% to 2.8%; <54 mg/dL fell from 1.3% to 0.3% (both p<0.02) A1c fell from 6.6% to 6%
The DBLG1 System Can Improve Blood Sugar Control in People with Poorly Controlled Type 1 Diabetes, Especially When the Initial Imbalance is High	DBL-US Study: RCT evaluating DBLG1 (n=88) vs. pump therapy (n=11) in 99 type 1s in France TIR +3.3 hours/day with DBLG1 vs. pump (p<0.001) Time <70 mg/dL was 12 min/day lower with DBLG1 than pump (p<0.01) Linear relationship between baseline TIR and TBR and amount of improvement with DBLG1 (p<0.001)
Impact of Delayed Prandial Insulin Boluses on Glucose Control in Patients on Advanced Technologies	Prospective evaluation of how timely pre-meal boluses are on AID vs. SAP and correlation with CGM-based outcomes Included 153 type 1s (122 on AID, 31 on SAP); averaged age 42, 9.2 delayed boluses Found delayed boluses are very common (about 1/5 of meals) and are associated with lower TIR, higher GMI, TAR, and time >250 mg/dL Delayed boluses directly correlated with fear of hypoglycemia while alone, at night, making unable to work, or losing consciousness (p<0.05 for each)
Glucometric Analysis in a Cohort of Pediatric Type 1 Patients Using Different Levels of Diabetes Technology from a Single Pediatric Center	Retrospective study evaluating four systems (Paradigm Veo, MiniMed 640G, MiniMed 670G, MiniMed 780G) in 188 children with diabetes (mean age 15, 10 years since diagnosis) in Italy 780G improved TIR +2 hr/day relative to 670G (71% vs. 79%); no change in hypoglycemia 780G showed a 66% reduction in Auto Mode exists relative to 670G (p<0.001) Those using 780G achieved >60% time in tight range Relatively similar outcomes with Veo and 640G; large improvements seen with 670G relative to Veo and 640G
Early Changes in Blood Glucose Control after Starting Auto-Mode with Different Hybrid Closed Loop Systems	Comparison of 780G (n=62) vs. Control-IQ (n=28) vs. DBLG1 (n=20) in adult type 1s one, three, and six months after starting Auto Mode Not randomized or matched so baseline differences may confound No significant differences in CGM metrics between Control-IQ and MiniMed 780G DBLG1 six-month Time in Range significantly lower than 780G or Control-IQ; GMI significantly higher than 780G
Treatment Satisfaction and Stress Level: Outcomes after One Month of Advanced Hybrid Closed-Loop (AHCL) Therapy in Latin America	Retrospective analysis of treatment satisfaction, stress, and glycemic outcomes in type 1s during first month of 780G use (n=9; age 31, baseline A1c 6.8%) Time in Range increased 1.7 hr/day from 70% to 77% (p=0.01) No significant change in hypoglycemia but was low at baseline 61% of participants reported high treatment satisfaction; DTSQ score improved 33 points

Glucose Sensors

Title	Details + Takeaways
The Impact Of Continuous Glucose Monitoring Data Loss On Glycemic Outcomes – Is 70% Of Data Sampling Over 14 Days Enough?	N=291, adults with type 1 diabetes using Dexcom G6 with a diabetes duration ≥2 years Created random data loss of 1-5 hours and removed CGM values until a range of 10-50% of data loss was achieved to investigate thresholds for amount of CGM data needed for accurate CGM metrics Optimal CGM wear time was dependent on the amount of missing data International consensus recommendation for 70% of CGM data over 14-days for accurate CGM metrics is sufficient given there are no large data gaps
Longitudinal Relationship Between Time In Range And HbA1c In A Real-World Clinical Practice Setting	N=542; adults with type 1 diabetes from the Barbara Davis Diabetes Center; average A1c of 7.5% and average Time in Range of 61% Researchers used longitudinal mixed models to estimate the correlations between real-world Time in Range and A1c at clinic visits 10% improvement in Time in Range associated with A1c reduction of 0.34% when estimating A1c at that clinic visit and A1c reduction of 0.2% when estimating A1c at subsequent clinic visit
The Association Between Body Mass Index (BMI) And Time In Range Among Young Adults With Type 1 Diabetes: Data From The T1D Exchange QI Collaborative	EHR data from T1D Exchange clinics in young adults (18-35 years old) from 2018-2020 analyzed (n=1,126) Patients with overweight or obesity (p=0.01), on public insurance (p=0.06), or of Hispanic race/ethnicity (p<0.006) were more likely to have lower Time in Range The majority of patients in the study (n=926) had Time in Range values <70% whereas under 25% (n=200) reported Time in Range >70%.
Reductions In HbA1c In Type 1 And Type 2 Diabetes With Flash Glucose Monitoring Are Sustained From 3-24 Months: A Meta-Analysis Of Real-World Observational Studies	Meta-analysis of 75 studies evaluating longitudinal FreeStyle Libre use (n=30,478 type 1s, n=28,063 type 2s) For people with type 1 diabetes, FreeStyle Libre use associated with A1c reduction of 0.5% after 3-4 months of use and 0.4% reduction after 4.5-7.5 months of use For people with type 2 diabetes, FreeStyle Libre use associated with A1c reduction of 0.45% after 3-4 months of use and 0.6% reduction after 4.5-7.5 months of use A1c reductions achieved at 3 months were maintained out to 24 months in adults with type 1 and 12 months in adults with type 2
Evaluation Of The Accuracy Of Continuous Glucose Monitoring Without Calibrations In Diabetic Patients On Intermittent Hemodialysis	N=20 adults with diabetes on hemodialysis (n=15 type 2s; n=4 type 1s; n=1 post transplantation) average age of 61 years old Investigating accuracy of Dexcom G6 Pro; participants wore one CGM for 10 days with CGM values compared to at home BGM and venous glucose during hemodialysis Compared to BGM, G6 Pro was found to have a MARD of 13.8% with 98.7% of values in the A/B zones of the Parker Error Grid demonstrating clinical accuracy Compared to venous glucose, G6 Pro was found to have a MARD of 14.3% with 100% of values in the A/B zones of the Parker Error Gird demonstrating clinical accuracy
Better Glycemic Control And Higher Use Of Advanced Diabetes Technology In Age Group 0-17 Yrs Compared To 18-25 Yrs With Type 1 Diabetes	N=725 people with type 1 diabetes between the ages of 0-25 in Sweden Children ages 0-17 demonstrated stronger glycemic management than young adults ages 18-25 with an A1c that was 0.3% lower on average (p<0.0001) Substantially higher percentages of pediatric patients (ages 0-17) used diabetes technology compared to young adults (ages 18-25) Pediatric patients saw greater reductions in A1c than young adults following care from a psychologist whereas young adults saw greater reductions in A1c following care from a dietician
Variation Of Time In Range	N=166 adults with type 1 diabetes wearing is-CGM shared eight weeks of CGM data; 84% on MDI, 26% on CSII therapy Four Time in Range intervals of two-weeks each were calculated for each patient Within-patient standard deviation for Time in Range was 6.3% and the chance of two two-week Time in Range values differing by more than 5% was 58%
First Data From The Ambulatory Glucose Profile Of Patients With Type 1 Diabetes At A Tertiary Hospital: A Retrospective Study	N=776 adults with type 1 diabetes in Spain using FreeStyle Libre 14-day or FreeStyle Libre 2 CGM At baseline, patients had an average Time in Range of 63% with an average mean glucose of 152 mg/dL, Time Below Range of 5%, and CV of 37.5% After ≥12 months, Time in Range improved to 64% (p=0.006), Time Below Range decreased to 3% (p<0.001), and CV decreased to 36% (p<0.001)
Effectiveness Of The Freestyle Libre 2 Flash Glucose Monitoring System On Diabetes-Self-Management Practices And Glycemic Parameters Among Patients With Type 1 Diabetes Using Insulin Pump	Prospective single center study of FreeStyle Libre 2 use among pediatric patients on pump therapy formerly using BGM (n=47, aged 13-21 years old) in Saudi Arabia At 12 weeks, Libre 2 data demonstrated an average Time in Range of 60% at follow-up, with Time Above Range of 33% and Time Below Range of 7% FreeStyle Libre 2 use was associated with improvements in subscales of the Diabetes Self-Management Questionnaire including glucose management (p=0.031), dietary control (p=0.048), physical activity (p=0.046), and health care use (p=0.024)
Barriers To Continuous Glucose Monitor Use In Youth With Type 1 Diabetes	N=136 teenagers with type 1 diabetes using CGM, average age of 15 years old with a diabetes duration of 6.6 years Top three reported reasons adolescents stopped using CGM were: (i) pain/discomfort; (ii) disliked wearing during school/exercise; and (iii) adhesive issues 73% of respondents initiated CGM use in the clinic and across all participants accuracy was noted as the area they would have liked to know more about prior to device initiation
Cost-Effectiveness Of Flash Glucose Monitoring With Optional Alarms In Swedish Adults With Diabetes And Impaired Awareness Of Hypoglycaemia, Using Intensive Insulin	IQVIA cost-efficacy analysis of FreeStyle Libre 2 use in Sweden among adults with type 1 diabetes and impaired hypoglycemia awareness; baseline characteristics derived from IN CONTROL and IMPACT studies FreeStyle Libre 2 was associated with higher up-front costs than BGM but was also associated with a reduction in severe hypoglycemic events and improved QALYs ICER for FreeStyle Libre 2 compared to BGM was ~$22,500/QALY FreeStyle Libre 2 use associated with reduction in severe hypoglycemic events of 11 events/year
Freestyle Libre 2 Vs Freestyle Libre 1 Comparison In Glycemic Control Outcomes In People With DM1	Comparison of FreeStyle Libre 14-day (n=36) glycemic outcomes 90-days after system initiation compared to FreeStyle Libre 2 (n=54) among adults with type 1 diabetes There was no significant difference in Time in Range between FreeStyle Libre 14-day and FreeStyle Libre 2 users at 58% and 61%, respectively Patients using FreeStyle Libre 2 experienced significantly less Time Below range compared to FreeStyle Libre 14-day users at 3.5% and 7.2%, respectively (p=0.002) Patients using FreeStyle Libre 2 experienced fewer hypoglycemic events compared to FreeStyle Libre 14-day users at 6.6 and 13.6 events, respectively (p<0.001)
Longitudinal Analysis Of Hypoglycemia In Rt-Cgm Users From Germany, Sweden, And The United Kingdom (2018-2020)	Investigation of real-world impact of Dexcom G5 or G6 use on hypoglycemia rates among adults with diabetes in Germany, Sweden, and the UK between 2018-2020 Across all geographies, G5 or G6 use was associated with a decrease of 0.27% in time <54 mg/dL 0-3 months after system initiation, and a decrease of 0.2% in time <54 mg/dL 3-6 months after system initiation G5 or G6 use was associated with a decrease of 0.5% in time <54 mg/dL 0-3 months after system initiation, and a decrease of 0.34% in time <70 mg/dL 3-6 months after system initiation G5 or G6 use was associated with an 8% increase in the percentage of patients achieving consensus targets for time <54 mg/dL <1% 0-3 months after system initiation G5 or G6 use was associated with a 6% increase in the percentage of patients achieving consensus targets for time <70 mg/dL <4% 0-3 months after system initiation
Impact On Quality Of Life And Glycaemic Outcomes In Adults With Type 1 Diabetes After Two-Year Use Of Flash Glucose Monitoring	N=229 adults initiating is-CGM use took fear of hypoglycemia and Quality of Life surveys at baseline, 6 months, and 24 months post CGM initiation Baseline A1c of 7.5%, average age of 47; 73% of patients on MDI, 27% of patients on pump therapy Quality of Life survey scores improved significantly at 24 months (p<0.01) No significant change in A1c, but Time in Range improved over the course of the study (p<0.05) Patients reporting hypoglycemia unawareness was reduced after 24 months (p<0.05)
Real-Life Evaluation Of The New Sensor Glucomen Day CGM	N=20 patients with type 1 diabetes; mean age of 33; 55% women; wore GlucoMen Day CGM for three months Prior glucose monitoring technologies included FreeStyle Libre (40%), Guardian sensor as part of a Medtronic AID system (25%), BGM (20%), Dexcom CGM (15%) There were no significant differences in A1c, fear of hypoglycemia, diabetes-related quality of life, or experience using glucose sensor between baseline (prior technology) and 3 months of GlucoMen Day CGM use 69% of participants “would accept to keep using the system.”
The Impact Of Glycemic Variability On The Relationship Between Hypoglycemia And HbA1c	N=542 adults with type 1 diabetes at the Barbara Davis Center for Diabetes using CGM; average A1c of 7.5%; average Time in Range of 61%; average age of 38 years old Coefficient of Variation <30% was associated with lower rates of Level 1 (<70 mg/dL) and Level 2 (<54 mg/dL) hypoglycemia Patients with A1c <6.5% and CV > 40% experienced the highest rates of both Level 1 and Level 2 hypoglycemia
Evaluating The Efficacy Of Freestyle Libre Flash Glucose Monitoring In The Older Population	N=36 older adults (≥65 years old) with diabetes using FreeStyle Libre CGM with an average of ≥8 scans/day 75% of patients met consensus guidelines for >50% Time in Range; 50% of patients met consensus guidelines for <10% Time Above Range 72% of patients met consensus guidelines for <4% Time Below Range, but only 17% met consensus guidelines for <1% time <54 mg/dL
Glucose Variability And Cardiometabolic Status In Type 2 Diabetes	N=30 adults with type 2 diabetes wore CGM for three months At baseline, women exhibited poorer glycemic management with higher glucose variability (GV) and mean amplitude of glycemic excursions (MAGE) compared to men at 52% vs 33% and 130 mg/dL vs. 87 mg/dL, respectively (p<0.05) After 3 months of CGM use, there was no significant difference in GV or MAGE between men and women, but both increased numerically for men from baseline Body weight was inversely correlated to both GV and MACE (r=-0.508 and r=-0.421, respectively)
Prospective Randomized Clinical Study Assessing The Impact Of Continuous Vs Flash Glucose Monitoring During A 8 Weeks Period In Children With Type 1 Diabetes	RCT evaluating use of rt-CGM (n=12) versus is-CGM (n=12) among pediatric patients with type 1 diabetes rt-CGM users experienced a significant reduction in Time Below Range from 12% at baseline to 7% after eight weeks (p=0.025), but there was no difference in Time Below Rang reductions between rt-CGM and is-CGM users rt-CGM users experienced a significant reduction in nocturnal hypoglycemia from 5% nighttime Time Below Range at baseline to 2% nighttime Time Below Range after 8 weeks (p=0.028) There was no significant difference in Time in Range between rt-CGM and is-CGM users
Monitoring The Incidences Of Allergic Contact Dermatitis And Other Skin Reactions Associated With Continuous Glucose Monitoring Systems	Literature review of association between CGM device use and contact allergy or dermatitis reactions among people with diabetes The most common skin-related reactions to CGM included Allergic and Irritant Contact Dermatitis Device placement and adhesive composition were both identified as common causes of CGM skin reactions
Use Of Continuous Glucose Monitoring Devices In Type 2 Diabetes To Track Disease Progression And Therapy Effectiveness	Model based analysis of CGM-derived type 2 diabetes disposition index compared to current standard of care oral minimal model disposition index for 100 virtual patients Standard oral minimal model included data on plasma glucose and insulin and C-peptide concentration following meals; CGM-based model used CGM data on glucose and glycemic variability following simulated meals The CGM-derived model correlated well to the oral minimal model  and were both able to significantly detect differences in diabetes status versus placebo (p<0.001)
The Link Between Microvascular Complications And A 10% Change In TIR In People With Type 1 Diabetes	Evaluation of the impact of a 10% change in Time in Range on microvascular complications over 2 years Adults with >10% decrease in Time in Range (n=65); adults with change (+/-) in Time in Range <10% (n=201); adults with ≥10% increase in Time in Range were compared Adults with type 1 diabetes using CGM who experienced ≥10% increase in Time in Range over 24 months were at increased risk for developing retinopathy (p=0.03), but not neuropathy or albuminuria Mean glucose was a strong predictor of developing retinopathy (p<0.001) and microalbuminuria (p=0.003)
Is There A Correlation Between Obesity And Glucometrics In People With Type 1 Diabetes?	N=42 adults with type 1 diabetes and obesity compared to n=83 non-obese adults with type 1 diabetes No statistically significant difference in A1c, Time in Range, or GMI between adults with and without obesity Time Above Range was numerically higher among adults with obesity at 41% compared to 37% among adults without obesity, but this difference was not statistically significant Average mean glucose was numerically higher among adults with obesity at 172 mg/dL compared to 164 mg/dL among adults without obesity, but this difference was not statistically significant
Updated Glucose Monitoring Communication (GMC) Survey	N=120 pediatric patients with type 1 diabetes (ages 8-17 years old) took the BGM Communication score with added questions related to CGM use termed the Glucose Monitoring Communication (GMC) survey Higher scores on the GMC indicate more negative perception of glucose monitoring technology Higher GMC scores among children were associated with more depressive symptoms (r=0.6, p<0.0001) Higher GMC scores among parents were associated with reduced CGM wear among children (r=-0.2, p=0.04)
Insights Into The Concept Of Safe Ramadan Fasting Using CGM-Derived Metrics	N=20 patients with CGM data collected ahead of and during Ramadan; data evaluated to identify changes in glycemic management Differences in patients achieving glycemic targets and those not were exacerbated during Ramadan for patients with type 1 and type 2 diabetes Patients with type 1 diabetes demonstrated better glycemic management during Ramadan compared to patients with type 2 diabetes Patients on insulin or sulfonylureas saw the largest declines in glycemic management while fasting
Performance Evaluation Of Subcutaneous Glucose Monitoring In Clinical Development—Optical Measurement Profusa Lumee Glucose Platform	Study investigating Lumee subcutaneous CGM in 16 patients with diabetes over three months 2156 paired sensor values; 81% overall measurements within 20/20 of BGM values; only 13% of values <54 mg/dL within 20/20 of CGM values Overall MARD from day 6-91 was 13.4%
Dynamics Of Cholesterol And Blood Pressure In 4,484 Adults With Type 1 Or Type 2 Diabetes After Initiation Of CGM	N=2,999 type 1s and n=1,485 type 2s; evaluating impact of CGM initiation on cholesterol and blood pressure Adults with type 1 diabetes using CGM saw a decrease in cholesterol after six months using CGM (p=0.001); adults with type 1 saw a slight increase in blood pressure after six months on CGM (systolic blood pressure p=0.016; diastolic blood pressure p=0.013) Adults with type 2 diabetes using CGM saw decreases in cholesterol (p<0.001) and systolic blood pressure (p=0.013) after six months on CGM
Psychosocial Outcomes In Young People With Type 1 Diabetes And Their Parents Using Glucose Sensors	Prospective cross-sectional study among pediatric patients with type 1 diabetes using CGM (n=70) and their parents/caregivers investigating fear of hypoglycemia and quality of life outcomes Baseline A1c of 7.5%; 31% on pump therapy at baseline; 63% on CGM at baseline Parents exhibited higher levels of fear of hypoglycemia compared to pediatric patients, especially compared to younger children (p<0.05) Among pediatric patients using is-CGM, higher scanning rates (≥10 scans/day) were associated with increased parental fear of hypoglycemia compared to parents of children with lower scan rates (p<0.05)
Characteristics Of People With Diabetes That Spend 15% Or More Time Below Range	N=27 patients with Time Below Range >15%; 55% type 2 diabetes, 30% type 1 diabetes, 15% hypoglycemia without diabetes diagnosis Overall study population exhibited high levels of glycemic variability with a CV of 42% Only 30% (n=8) of study population had a glucagon prescription
Patient Perspectives On Factors Affecting CGM Accuracy, Utility And Decision Making In People With Diabetes Within The T1D Exchange Community	N=605 current and former CGM users with either type 1 or type 2 diabetes from the T1D Exchange registry completed survey on CGM use 49% of study population had temporarily discontinued CGM use at one point in diabetes management Top five reasons for stopping CGM use included: (i) inaccurate readings; (ii) sensors falling off before end of sensor period; (iii) skin irritation; (iv) too many alarms; (v) too many false alarms 90% of survey population felt sensors were accurate, but 42% felt that accuracy varied between sensors
Percentage Of Time In Range That Defines Good Glycemic Control In Patients With Type 1 Diabetes	N=118 adults using AID; 94% type 1s; 47% overweight or obese Moderate inverse association between Time in Range and A1c (r=-0.54, p<0.001) Time in Range >75% was more predicative of an A1c <7% compared to Time in range >70% with a sensitivity of 70% and specificity of 67%
Performance Of A Novel, Minimally-Invasive Wearable CGM (K’Watch System) In Diabetic Patients: A Prospective Clinical Trial	N=4 adults with diabetes wore K’Watch microneedle CGM system; system lasts for seven days MARD of 18% compared to commercially available CGM Participants reported no adverse reactions at microneedle insertion site
Improved Time In Range Associated With A Continuous Glucose Monitoring System’s Automated Notification And Retrospective Analysis Features	Patients with type 1 or type 2 diabetes using Dexcom G6 and Dexcom Clarity In type 1s, use of Dexcom Clarity was associated with higher Time in Range at 64% compared to among patients who did not use Clarity with an average Time in Range of 56% In type 2s, use of Dexcom Clarity was associated with higher Time in Range at 65% compared to Time in Range of 58% among those not using Clarity
Use Of Flash Glucose Monitoring (FGM) System In Older Type-2 Diabetics	N=204 adults with type 2 diabetes using is-CGM and MDI (average age of 70 years old) Baseline A1c of 8.9% CGM use associated with average A1c reduction of 1.4% to 7.5% after two years (p<0.05)
The Assessment Of Impact Of Real-Time Continuous Glucose Monitoring On People Presenting With Severe Hypoglycaemia (Air-CGM) Study	N=30 adults with type 1 diabetes; evaluating CGM use immediately following hospitalization for severe hypoglycemia RCT; control group wore blinded CGM, intervention wore Dexcom G6 CGMs CGM use was associated with fewer severe hypoglycemic events compared to control with zero and four events for each arm, respectively (p=0.04)
Intermittently Scanned Continuous Glucose Monitoring In The Postoperative Management Of Hyperglyceamia After Pancreaticoduodenectomy – A Comparison With Point-Of-Care Capillary Testing	N=88 patients undergoing pancreaticoduodenectomy wore FreeStyle Libre CGMs to monitor inpatient glycemic levels post operatively 2,346 pairs of CGM and BGM data collected; MARD of 17.8% POC BGM glucose levels were consistently higher than CGM values
Continuous Glucose Monitoring (CGM) In Patients Following Simultaneous Pancreas And Kidney Transplantation	N=43 adults with type 1 diabetes who underwent simultaneous pancreatic and kidney transplant 54 periods of 72-hour CGM wear were performed post operatively out to 4.7 years post-transplant 99% of CGM readings were in range

Insulin pumps and other insulin delivery devices

Title	Details + Takeaways
Delivery Accuracy and Occlusion Detection Time of Sigi, a Novel Patch Pump to be Used with Standard 1.6 mL Insulin Prefilled Pump Cartridges	Evaluation of a prototype of Sigi patch pump in terms of basal rate, bolus accuracy, occlusion detection time Very little basal rate deviation over 72 hours (-0.3±2%); same true of meal boluses (1.8±2.5% at volumes of 0.2 U and 0.3±2.4% at volumes of 1.0 U) Mean occlusion detection time of 29 minutes w/basal rate of 0.1 U/h and 10 minutes w/basal rate of 1.0 U/h These accuracies are similar to those of commercially available pumps; shorter occlusion detection time than currently available pumps
Self-Management of Insulin Pump Settings and Data Upload are Associated with Lower HbA1c	Assessment of the correlation between PWD-driven management of pump settings and data uploads w/A1c Used EHR + survey data from 770 adult pumpers in Denmark (age 49, mean A1c 7.3%) On uploading pump data (n=747): 50% never did, 36% did ≥2-5x/year, which was associated with lower A1c than never uploading (p=0.049) On users adjusting settings (n=749): 28% never adjusted, 32% adjusted, which was associated with lower A1c than non-adjusters (p<0.001); basal rate + I:C ratio most frequently adjusted by users
Continuous Subcutaneous Insulin Infusion is Associated with Better NAFLD Indices in Patients with Type 1 Diabetes	Cross-sectional retrospective analysis evaluating NAFLD indices in type 1 adults based on pump (n=260) vs. MDI (n=399) Several baseline characteristics differed significantly between groups (age, gender, duration of diabetes, waist circumference, insulin/kg, plasma triglyceride, HDL cholesterol); however, controlled for these in analysis Pump use associated with lower HSI and FLI values, independent of glycemic control and all other potentially confounding factors except gender Predicted that lower glycemic variability and peripheral hyperinsulinemia are likely factors associated with pump’s influence on NAFLD prevalence
Time-In-Range Patterns around Clinic Visits among Patients with Type 1 Diabetes Using a Smart Insulin Pen	Post hoc analysis of single-arm prospective proof-of concept study evaluating patterns of glycemic control and injection behavior around clinic visits among adult type 1s using NovoPen 6 and CGM (n=87); evaluated the 28 days before and after visit TIR generally increased over the study period, but peaked during the first 14 days post-visit and then decreased for the subsequent 14 days Missed bolus doses/day decreased over study period but generally increased after each visit before decreasing again
Use Of Decision Tree Modeling To Identify Factors Associated With Optimal Time-In-Range (TIR) During InPen™ Smart Insulin Pen And Continuous Glucose Monitor (CGM) Use	Analysis of InPen CGM data from 1,710 adult type 1s starting InPen between July 2020 and June 2021 The top variables associated with Time in Range were: (i) frequency of missed meal doses, (ii) diabetes duration; (iii) bolus count; and (iv) correction frequency The highest Time in Range (81%) was observed in those with diabetes duration between one and four years and ≤21% missed doses Data highlight that missed meal boluses were a major modifiable factor associated with Time in Range
Comparison Of SAP vs MDI with CGM in Very Young Children with Newly Diagnosed Type 1 Diabetes: Clinical Effectiveness and Long-Term Benefits	Nine-year prospective assessment of SAP (n=24) vs. MDI/CGM (n=30) at diabetes onset in very young children (n=54; age <4) SAP users saw -1.6% lower A1c than MDI/CGM users at nine years (6.7% vs. 8.3%) from similar baseline values (10.4% vs. 10.1%) Average Time in Range over study period +4.8 hr/day higher among SAP users than MDI/CGM users (65% vs. 48%)
Long-Term Cost-Effectiveness Using The Smart Insulin Pen Cap Insulclock® In Relation To The Standard-Of-Care Treatment In Type 1 Diabetes Population In Spain	Microsimulation analysis assessing cost-effectiveness of Insulclock smart pen cap system compared to Standard of Care in Spain Clinical data pulled from RCT (n=21 type 1s on basal-bolus therapy and CGM); app info and alerts associated with +1.9 hours/day Time in Range (p=0.026), translating to -0.6% A1c improvement Unitary costs derived from Spanish literature and official tariffs for healthcare services; incorporated 3% annual discount rate and 2% inflation rate Insulclock was more effective than SoC and associated with less total cost (-36 EUR/patient) and average savings of 996 EUR per year of life
Fear of Hypoglycemia between People with Type 1 Diabetes on Insulin Pump Versus Basal Bolus Insulin Therapy	Assessment of the correlation between fear of hypoglycemia and pump therapy (n=40) vs. MDI (n=55) in type 1s (n=95; age 47, 7.7% A1c; 57% used CGM) Used the Hypoglycemia Fear Survey (HFS) questionnaire Those on pump therapy had a lower HFS score than those on MDI (p=0.03); no difference based on CGM use HFS score was positively correlated with diabetes duration (p=0.01) and negatively correlated with dyslipidemia (p=0.002) and total daily insulin units (p=0.001)
Prevalence of Emotional Stress among People with Type 1 Diabetes and Insulin Pump versus Basal Bolus Insulin Therapy	Assessment of the correlation between diabetes-related emotional burden and pump therapy (n=40) vs. MDI (n=55) in type 1s (n=95; age 47, 7.7% A1c; 57% used CGM) Used the Problems Areas in Diabetes Scale (PAID) questionnaire PAID was related to BMI (RR=1.12), duration of diabetes (RR=1.08), A1c (RR=1.5), presence of coronary artery disease (RR=0.04), CKD (RR=4.96), use of insulin pump (RR=0.14), CGM (RR=4.13), and hypoglycemia events (RR=1.01) Use of insulin pump associated with lower diabetes-related emotional distress, whereas use of CGM associated with higher diabetes-related emotional distress
Perception of Quality of Life, Degree of Satisfaction and Glycemic Control in Users of YpsoPump	Survey-based evaluation of treatment satisfaction and QoL with YpsoPump in type 1s on Spain (n=26, average age 40, 21 years of diabetes, used YpsoPump for average 14 months) Self-reported A1c fell from 8.3% before YpsoPump to 7.1% after 6-12 months of use (p<0.01) A majority of participants self-reported improved glycemic control and decreased episodes of hypoglycemia with YpsoPump (84% and 80%, respectively) Diabetes Impact and Device Satisfaction Scale: satisfaction = 9.4; main benefits perceived included more flexibility, better postprandial control; main barriers perceived included visibility of diabetes and alarm burden 12% reported technical problems
Glycemic Control in Children and Adolescents with Type 1 Diabetes (T1D) from The Onset: Continuous Subcutaneous Insulin Infusion (CSII) vs. Multiple Daily Injections (MDI)	Retrospective analysis comparing efficacy and safety of SAP (n=28) vs. MDI/CGM (n=16) at diabetes onset in children/adolescents with type 1 (n=44, average age 9) Those on pump had lower basal insulin requirement than those on MDI (p<0.01) Similar glycemic metrics (both A1c and CGM metrics) at three and six months with SAP vs. MDI/CGM At 12 months, TIR trended higher in SAP group vs. MDI/CGM group but not significant (70% vs. 55%)
Clinical Characteristics of Patients with Type 1 Diabetes Mellitus who Follow Insulin Pump Therapy at the Marqués De Valdecilla University Hospital	Retrospective study of all type 1s who used pump therapy for at least one year since pumps began use at this hospital Included 127 participants: 58% women, mean age 46 years, diabetes duration of 27 years, A1c at start of follow-up was 8%, used pump for average eight years Reason for pump indication: 43% poor control, 30% repeated hypoglycemia, 9% glycemic variability, 4% dawn phenomenon Mean A1c decrease of 0.5% with pump therapy (8% to 7.5%); greater improvement if always used bolus calculator vs. occasionally (7.4% vs. 7.9%); 43% had an A1c ≥8% at baseline
The Impact of Continuous Insulin Infusion and SGLT-i On Glycemic Control of Night-Shift Workers with Type 1 Diabetes	Retrospective analysis the combined use of pump/SGLT-2 (n=5) vs. pump alone (n=23) in night shift workers with type 1 (n=28) using FreeStyle Libre data and A1c Participants averaged age 33, diabetes duration of 21 years, using pump for 26 years, A1c before using pump 8% Those on pump + SGLT-2 had +2.2 hr/day Time in Range vs. those on pump alone (69% vs. 60%, p=0.01) and -0.3% lower GMI (6.8% vs. 7.1%, respectively, p=0.02) No significant differences in hypoglycemia or TBR; rates of DKA not reported

Clinical Decision Support Systems/Advisors

Title	Details + Takeaways
Evaluation Of An Insulin-Carbohydrate-Ratio Adjustment Algorithm Performed On Real Patients Datasets	Retrospective study in adults (n=14, mean postprandial glucose 154 mg/dL, 2.5 meals/day) and children (n=15, mean postprandial glucose 114 mg/dL, 3.8 meals/day) evaluating novel insulin-carb-ratio (ICR) algorithm All calculated ICRs within physiological level and eventually converged toward physician reference ratios given by doctors (by day nine for children and day 14 for adults) CGM and insulin-on-board data facilitated ICR calculations; algorithm struggled to account for postprandial hypoglycemia or hyperglycemia
Novel AI-Based Algorithm To Detect And Reconstruct Meal Real Time Using CGM Data	Retrospective, observational study (n=14, mean age 51, baseline A1c 7.1%) evaluating novel meal detection algorithm Trained model using all data except from one participant, from whom performance is evaluated; process repeated for each participant then averaged Detection time range (period given to model to detect a meal after it occurred) of 60 mins led to area under ROC curve (AUC) of 92; detection time range of 45 mins led to AUC of 89 Preliminary data suggest algorithm is highly promising at detecting meals from real-life CGM data
Impact Of A Digital Therapeutic Platform On Weight Loss And Diabetes Self-Management	Retrospective one-year study of Dario Health program members (n=715, baseline BMI ≥30 kg/m2, 51% male, 80% type 2s) investigating impact of digital health intervention on weight loss and diabetes self-management Whole population reduced weight on average (p<0.05); nearly 2/3 of participants reduced weight by an average 7.4% (p<0.05) and reduced BMI by an average 2.8 kg/m2 (p<0.05) Over 30% of participants achieved weight loss of ≥5% at year one (p<0.05); subset with baseline BMI ≥35 kg/m2 (n=237) achieved ≥5% weight loss at year one (p<0.05) Subgroup with baseline mean glucose >180 mg/dL reduced weight by mean 4.9%, average glucose by 16.1%, and high readings ratio by 38% at year one (p<0.05)
The My Friend Diabetes Carbohydrate Bolus Calculator: User Experiences	User experience study (n=165, 35% type 1s, 53% female, mean age 13) of bolus calculator with nutrition database that facilitates carb counting Calculator uses glucose value, carbs (grams), insulin sensitivity, and insulin:carb ratio to make calculations 130 participants (~79%) said app improved diabetes management and carb/dose calculations; more confident in diabetes because of app, and less restricted by food
Use Of Diappymed® App For Carb&Bolus Counting Results In Improved Post-Meal Glucose Control In Patients With Type 1 Diabetes In A Randomized Control Study	Randomized controlled crossover study validating accuracy of DiappyMed app for carb and bolus counting (n=28, 33% male, mean age 43, baseline A1c 7.5%, all using BGM) Two randomized phases (each one month) without washout (either using app, or not [control]) App users spent +1.4 hours/day between 80 and 180 mg/dL compared to control group (p<0.0002), completely due to reduction in TAR 13 participants (of 24) said they would use it daily; 10 participants (of same 24) said they would use it many times a week
Cardiovascular Risk Assessment In People Living With Type 1 Diabetes From The Renaced-Dt1 Mexican Registry	Model classifying Mexican individuals in RENACED-DT1 registry according to CV risk using Steno-T1D score; gauged risk of microvascular complications Included type 1s (n=1,718) with no prior CVD events; risk categorization based on NICE guidelines (86% low risk, 9% middle risk, 5% high risk); microvascular complications assessed via logistic regression High-risk CV patients had high risk of retinopathy (OR=8.5), neuropathy (OR=5.4), and CKD (OR=8.3); model suggests routine CVD risk assessment can highlight risk factors for microvascular disease
Retrospective Study Correlating Self-Monitoring Blood Glucose (SMBG) Values With HbA1c	Retrospective analytic model correlating A1c with BGM in type 1s (n=20, >18 years old, >20 pre-meal and bedtime BGM checks/week) In four participants, >70% of BGM readings were in Range, translating to mean A1c of 6.9%; in eight participants, 65%-70% of readings were in Range, translating to mean A1c of 7.2%; in four participants, 60%-65% of readings were in Range, translating to mean A1c of 7.5%; in two participants, 55%-60% of readings were in Range, translating to mean A1c of 8.1%; in two participants, 45%-50% of readings were in Range, translating to mean A1c of 8.4%
Evaluation Of An Enhanced Care Intervention Using An Artificial Intelligence-Guided Decision Tool In Children And Emerging Adults With Type 1 Diabetes	Prospective study assessing feasibility, acceptability, and efficacy of DreaMed Advisor Pro within Glooko to provide insulin recommendations between visits in children and emerging adults with type 1 diabetes Participants (n=95) had type 1 for at least six months, aged 7-26 years (mean 13.5), and wore Omnipod insulin pump and Dexcom CGM Baseline A1c/GMI ranged from 7% to 13% (mean 7.8%); baseline Time in Range of 45%; 99% white; 76% privately insured; 51% had household income ≥$100,000 No change in Time in Range, Time Below Range, or A1c at three months(n=59) or six months (n=33); but, feedback was largely positive (negative feedback was related to interoperability)

Digital Health and Telehealth

Title	Details + Takeaways
Engagement And Weight Loss From A Coached Digital Support Program In People With Type 2 Diabetes	Prospective data analysis of 52,156 in-app actions (n=400 type 2s) over 16 weeks from Noom Weight loss program 65% of users with available weight lost weight Engaged users (those performing ≥one app action in 8 of 16 weeks) lost 10 lbs (-5% from baseline) compared to -2 lbs (-1% from baseline) for those non-engaged (4% difference, p<0.05) Results suggest that a coached, digital support program for type 2s promotes weight loss via holistic lifestyle intervention, with participants who showed greater engagement exhibiting clinically meaningful weight loss
Glycemic Control Of Patients With Diabetes In Russia Who Were Using The Contour®Plus One BGMS With Contour®Diabetes App	ANOVA analysis comparing A1c from 601 patients who used the Contour Diabetes App (CDA) from January 2019 to September 2020 Mean A1c decreased significantly from 7.6% to 6.3% (p<0.0001) Compared to first 30 days of system use, there was approximately a threefold decrease in the estimated frequency of very high blood glucose (≥250 mg/dL, OR=3.6) and high blood glucose (≥180 mg/dL, OR=2.8) after >180 days of using CDA This real-world study showed significant improvement in self-reported A1c and increased likelihood of readings in Range among Russian diabetes patients using CDA system
Real World Evidence Of Improved Glycemic Control In People Using The OneTouch Verio Reflect® Glucose Meter With The OneTouch Reveal® Mobile Application	Analysis of OneTouch Verio Reflect glucose readings and app analytics from 4,154 type 1s and 13,623 type 2s; comparing data from first 14 days vs. 14 days before 90-day time point Time in Range increased +1.9 hours/day from 58% to 66% in type 1s and by +2.6 hours/day from 72% to 84% in type 2s; Time Above Range improved by -2.2 hours/day (from 37% to 29%) in type 1s and by -2.6 hours/day (from 26% to 15%) in type 2s
Telemedicine Follow-Up Of Adolescents With Type 1 Diabetes Mellitus With One Touch Reveal® Mobile App	26-week study assessing safety and efficacy of telemedicine app for care of adolescents (n=56 type 1s; aged ≥14 to 18 years; mean age 16 years with mean diabetes duration 7 years) A1c decreased significantly by 3rd and 6th months of the study: -0.3%; p=0.005 and -0.5%; p<0.001, respectively Participants saw a 1.3 hour/day increase in Time in Range (p=0.016) and a 3.1% decrease in %CV (p=0.015) Significant overall improvement in QoL reported by participants (p=0.008)
Which Diabetes App Features Improve Glycemic Control In Type 2 Diabetes? A Scoping Review	Meta analysis of seven articles to explore which specific features in apps for people with T2D improve glycemic control Three features showed significant effect on glycemic control: (i) monitoring and feedback system; (ii) gamification; and (iii) diabetes education Suggests diabetes apps may support glycemic control in T2D and should include educational components, gamification, and a feature for sharing data with HCPs to improve glycemic control
Initiating CGM Over Telehealth Is Well Accepted By Parents Of Newly Diagnosed Youth With T1D	Focus groups and interviews conducted with 16 parents of children who started CGM virtually within 30 days of diagnosis Benefits of virtual care included convenience and ease of scheduling; user friendliness; and being in the comfort of home, especially for young children Downsides included long visits with “information overload”; the challenges of CGM initiation; reported anxieties about “getting it wrong”; and less perceived support due to not having clinic staff in room Over half of parents preferred the virtual visit; participants suggested clinic should offer choice of in-person or virtual options in future
Improvement In Dietary Behavior And Glycemic Control For People With Type 2 Diabetes On Diabefly® Digital Therapeutics Platform	De-identified data from 69 people with T2D (mean age 45 years, 44% female) using Diabefly platform After 90 days on the platform, mean calorie intake was significantly reduced by 16% (p<0.001) Daily carb and fat intake were reduced by 18% (p<0.0001) and 16% (p<0.0001), respectively Participants saw a significant mean reduction in A1c (-1.9%, p<0.0001), body weight (-3.2 kg, p<0.0001), and BMI (-1.1 kg/m2, p<0.0001) After 90 days on the program, significant changes in dietary behavior were observed, which were associated with improved glycemic control among people with T2D
Utilizing Primary Care Providers To Advance Diabetes Technology Equity: Findings From Project ECHO T1D	Analysis of 70 PCPs who completed surveys following participation in pilot ECHO T1D project, which sought to empower PCPs to manage T1D using diabetes tech From baseline to post-intervention survey, PCPs reported improved confidence in their ability to serve as a T1D provider for their community (+0.4, p<0.0001), manage insulin therapy (+0.5, p<0.0001), address psychosocial barriers (+0.5, p=0.0001), and identify symptoms of diabetes distress (+0.7, p<0.0001) PCPs reported improved confidence in all aspects of diabetes tech use, especially determining which patients would benefit from CGM (+0.7, p<0.0001), prescribing CGM (+0.5, p<0.0001), utilizing and interpreting CGM data (+0.6, p<0.0001), and determining which patients would benefit from insulin pump therapy (+0.6, p<0.0001)
A 12-Month Randomized Controlled Study Evaluating Telemedicine As A Partial Replacement For Standardized Care	12-month RCT comparing telemedicine vs. standard of care in 75 type 1s aged 2-17 years in Sweden Telemedicine conferred a greater QALY of 1.5 compared to 1.4 for the control group; telemedicine was cost effective with ICER 15,600 SEK ($1,551 USD) per gained QALY (low cost <100,000 SEK / $9,942 USD) Telemedicine led to non-inferior changes in A1c, TIR, TAR, and TBR and was related to significantly higher independency, higher treatment satisfaction, and higher self-efficacy The authors conclude that it is cost-effective to use telemedicine in addition to standardized visits among pediatric patients with type 1
Diabetes Device Data In Virtual Clinic Visits: A New Health Disparity?	Observations from 946 telehealth encounters of individuals <23 years old with type 1 diabetes in Midwest USA from March 2020 to November 2021 Only 53% (n=383) of participants had device data uploaded before virtual visits Mean A1c (9.5% vs. 8.5%, p<0.001), was lower and mean TIR (36% vs. 45%, p< 0.001) was higher in those that had uploaded/streamed data before their clinic encounter Those with a longer diabetes duration, with public insurance, or who self-identified as Black/African American were less likely to have data available at clinic visit
Training, Initiation And One-Year Follow-Up Of Insulin Pump Therapy By Telemedicine	Remote training of four patients initiating insulin pump therapy via three to four Skype sessions, with emphasis on self-management and technical aspects of pumps As early as two weeks after pump initiation, Time in Rane was over 90% in three of four participants During one-year remote follow up, all participants maintained satisfactory Time in Range and %CV, with a limited amount of hypoglycemia These findings support telemedicine for remote training, initiation, and follow-up of insulin pump therapy with beneficial effects in glycemic management
Detecting The Risk Of Type 1 Diabetes Through Clustering Of Data Collected During A Self-Administered CGM-Based Home Test	Data examined from CGM in 55 healthy relatives of type 1s having either zero, one, or ≥two islet antibodies (mean age 25, baseline A1c 5.3%, BMI 24 kg/m2) Nine clusters were identified in the CGM traces of 3-hours duration around each mixed meal tolerance test (MMTT) A statistically significant relationship between the nine clusters originating from the 44 (1 AB), 57 (2+ AB), and 54 (negative) autoantibodies (p=0.0004) was observed The data suggests that a new self-administered clustering technique based on home CGM traces in response to a MMTT is a relevant method for assessing T1D risk
Clinical Practice Insights Delivering Care During COVID In Five European Countries Utilizing A Professional Diabetes Management Ecosystem	Analysis of survey data from 22 HCPs from 20 European institutions who used OneTouch Reveal (OTR) system for BGM data Remote consultations increased 46% during COVID. HCPs reviewed OTR Pro data during (45%) or before (25%) consultations, every 3 months (20%) or every 2 weeks (5%) 55% of HCPs said going forward, OTR Pro would become their standard of care, 30% for current or new patients, 10% during face-to-face visits, and only 5% returning to face-to-face consultations without OTR Pro HCPs ranked the top benefits of OTR Pro as: “allows me to make treatment/therapy decisions,” “helps me schedule consultations/reminders,” and “access 24/7 to status of my patients”
Assessment Of Blood Glucose Readings Of People With Diabetes In Australia Who Were Using The Connected Contour® BGMS And Contour®Diabetes App	Analysis of anonymized glucose readings from 7,047 Contour Diabetes App (CDA) system users (>60% T2D) within first 30 days (baseline) and between 180-210 days of system use From baseline, participants saw an approximately sixfold reduction in the estimated frequency of very high blood glucose (≥250 mg/dL; mean OR=5.6), high BG (≥180 mg/dL, mean OR=6.5) The reduction in low glucose readings (≤70 mg/dL) was less pronounced (mean OR=1.2), and the reduction in very low glucose readings (≤54 mg/dL) was even less prominent (mean OR=1.2) Results support that monitoring BG levels and actively using CDA for at least 6 months may lead to improved glycemic control
Effectiveness Of Lilly Connected Care Program In Improving Glycemic Management Among Chinese Type 2 Diabetes Patients: A Retrospective Real-World Study	Analysis of 303 pairs of propensity-matched Chinese adult type 2 patients from January 2015 to January 2020 The A1c reduction during four-month follow-up was significantly larger in the Lilly Connected Care Program (LCCP) group than non-LCCP group (mean=2.2 vs. 1.7; p=0.003) LCCP group had higher proportions of patients with A1c reduction ≥1% (69% vs. 57%, p=0.003) and ≥0.5% (75% vs. 68%, p=0.038) Proportions of LCCP and non-LCCP group reaching target A1c levels were 29% vs. 20% for A1c ≤6.5% (p=0.011) and 42% vs. 36% for A1c <7.0% (p=0.114), respectively
Using Geocoding Technology To Determine The Cross-Sectional Association Between Living In A High Crime Area And Diabetes	Cross-sectional study of n=2,923 veterans ≥65 years old residing in Miami-Dade or Broward Counties Participants were 97% male, 61% Caucasian, 82% Non-Hispanic, 51% frail, 54% married, and had a mean BMI of 30 kg/m2 Using multivariate MLR, a higher crime index was not associated with poorly controlled diabetes (OR=1.0, p<0.648) Further research in more diverse populations and geographic areas is needed to clarify the associations
Co-Designing A Digital Health Platform For Delivering A Complex Intervention To People With Type 1 Diabetes And Disordered Eating	Nine participants (7 type 1s and 2 HCPs) took part in the STEADY experience-based co-design app-workshop The app includes psychometric questionnaires, diabetes and thought diaries, symptom tracking, goal setting, two-way messaging function, and a library of worksheets Participants favored a personalized and easy-to-use app to complete their CBT exercises and psychometrics, keep logs, review graphs tracking diabetes and mental health parameters, write notes, set reminders, and communicate with trial clinicians, without focusing solely on physical health or body weight Future suggestions included peer support and integration with other apps
Glucose Reduction In Employees With Diabetes After Long-Term One Drop Use	Retrospective study evaluating change in weekly average glucose and glucose above target (A1c ≥7%) among 97 employer-sponsored people with diabetes who participated in One Drop’s multicondition program for ≥six months 50% of participants were male, 62% were 40-60 years old, and One Drop use ranged from six to 28 months Those using One Drop for ≥six months experienced a significant 0.9% reduction in A1c Those who participated for 1 year or more (n=29) similarly experienced a significant 1.2% reduction in A1c
Usage Of Metabolic Tracker Device (Lumen®) Improves Metabolic Control In Adults With Prediabetes	12-week single-arm intervention study among 27 adults with prediabetes who received Lumen device as a daily tool Results revealed a significant decrease in body weight (-6 kg, p<0.001) due to significant decline in body fat % (-2.9%, p<0.001) and waist circumference (-6.2 cm, p<0.001) Significant reductions were also observed for A1c (-0.3%, p<0.001), triglycerides (-0.5 mg/dL, p<0.001), and systolic blood pressure (-5.0 mmHg, p<0.05)
Digital Meal Logging: Meal Composition Of Breakfast, Lunch, And Dinner Predicts Weight Loss In 11758 Patients	Retrospective analysis of ~1.8 million food logs and weight data from 11,758 people with obesity that participated in blended-care weight loss interventions The majority of participants were female (n=8,194), with mean baseline BMI 37 kg/m2 and mean relative weight change at week 12 of -3.5% For breakfast, four meal compositions (MCs) yielded positive effects on weight loss (“cereals & dairy”, “fruits”, “fruits & cereals & dairy”, “fruits & bread & dairy,” all pd (“probability of direction”) >0.95) For lunch, 3 MCs yielded positive effects (“fruits”, “soup & vegetables”, “fish & salad”, all pd>0.95) For dinner, 7 MCs proved beneficial (all lunch MCs; “fish & vegetables”, “red meat & vegetables”, “salad & white meat”, “white meat & vegetables”, all pd>0.95) The results confirm the well-established influence of lower-carb MCs on weight loss
Qualitative Evaluation Of A Telemonitoring Solution For People With Type 2 Diabetes On Insulin: A Pilot Study In Preparation For The DiaMont Trial	Qualitative study of five participants (ages 55-74, diabetes duration 0-20 years) who used a telemonitoring solution (activity tracker, CGM, and 2 smartphone apps) for four weeks Five themes emerged from post-study interviews: (i) information overload; (ii) user manuals; (iii) activity tracker and app use, (iv) CGM and app use; and (v) the overall experience Overall, participants found that the telemonitoring solution contributed positively to their glycemic control, self-management, and wellbeing However, the heavy load of information and introduction to several devices at inclusion was overwhelming; some participants also experienced problems with the smartphone apps (unwanted alarms and poor usability)
The Effects Of The COVID-19 Pandemic On People With Type 2 Diabetes Using One Drop	Survey to describe self-reported impact of pandemic on life areas, preventative care appointments, and health management behaviors among 171 type 2s Mean age was 55; participants were 55% male, 70% white, 95% insured, 89% COVID-vaccinated, and 85% had not had COVID Due to the pandemic, 94% reported negative life impacts, 63% reported missed/delayed preventative appointments, and 76% reported impacts on diabetes-related health behaviors Missed/delayed appointments were dental cleanings (42%), A1c or blood pressure checks (35%), and primary care visits (25%); for health behaviors, physical activity (68%), vegetable intake (32%), and carb tracking (37%) were most impacted For those reporting impact, higher levels of impact were associated with higher levels of diabetes distress and A1c in each impact area (p<.05); however, missed appointments were unrelated to A1c
Patients’ Perspectives On Using Video Consultations For Type 1-Diabetes Treated With Insulin Pumps In The Outpatient Clinic	Qualitative exploration of patient perspectives on using video consultations for type 1 management via semi-structured interviews of nine type 1s on pump therapy Perceived facilitators include: (i) less time spent on visits; (ii) easier to find time for minor issues; (iii) common understanding via screen sharing; and (iv) visual confirmation when changing insulin pump settings Perceived barriers include: (i) possible technical issues with either data uploading or conducting video consultation; (ii) performing a physical examination; and (iii) difficulty in building patient-provider relations Overall, participants believed that video consultations would be advantageous in the outpatient clinic, though they would prefer some in-person contact, especially to ensure patient-provider relationships

Time in Range and Beyond A1c

Title	Details + Takeaways
Utilization Of Time In Range In Real World Varies By Type Of Diabetes	Online survey of 985 adults with diabetes and 44 caregivers of children and/or adults with diabetes to assess the role of Time in Range when setting health goals and monitoring glucose data For the 86% of respondents who set personal diabetes health goals, consideration of Time in Range among one’s top three goals was higher for those with T1D vs. T2D (47%, n=175 vs. 15%, n=687) Among respondents who set Time in Range goals, 72% of those with T2D (n=109) reported discussing Time in Range with their HCPs compared to 89% of those with T1D (n=83) Time in Range was more likely to be identified as the most important diabetes metric by respondents who used CGMs (34%, n=322) and those with T1D (30%, n=201) compared to CGM non-users (2%, n=676) and respondents with T2 (8%, n=797)
Correlation Of HbA1c and Time In Range (Single Ambulatory Glucose Profile Report) in Persons With Type 2 Diabetes	Retrospective observational study of 350 adults with T2D who were given single-time FreeStyle Libre Pro 2 AGP CGM Average A1c values were 7.2% among patients whose Time in Range was >70% Those whose Time in Range was between 50% to 70% had an average A1c of 7.8% Patients whose Time in Range was <50% had an average A1c of 8.5% This data suggests that a single 14-day period of CGM is sufficient to predict glycemic control over a three-month period, as Time in Range correlates well with A1c
Sleep Quality And Its Relationship To Continuous Glucose Monitoring Metrics	This study assessed sleep quality in 168 women with gestational diabetes (GDM) who wore masked Medtronic iPro2 CGM for 6 days Participants had a mean age of 33 years, BMI of 31, and mean gestational age of 31 weeks 67% of participants had a Pittsburgh Sleep Quality Index (PSQI) score >5, indicating poor sleep quality Each unit increase in PSQI score was associated with a 0.030 mmol/L increase in mean glucose (ß 0.030; 95% CI 0.005-0.056; p=0.019) and a 0.026 mmol/L increase in SD glucose (ß 0.026; 95%CI 0.011-0.039; p<0.001) Participants with a higher PSQI score spent significantly less Time in Range (OR 0.90; 95% CI 0.82-0.99; p=0.037) and more Time Above Range (OR 1.15; 95%CI 1.04-1.27; p=0.008)
Relative Contribution Of Fasting Plasma And Postprandial Glucose To Hba1c And TIR In People With T1d On Basal-Bolus Insulin Therapy	Retrospective analysis of PRONTO-T1D (n=1,222) and its 14-day CGM study (n=269) Overall, a 1 mmol/L change in FPG or PPG was associated with significant changes in A1c and TIR for all treatment groups On ultra rapid-acting insulin lispro (URLi), a 1 mmol/L reduction in FPG and PPG was associated with 0.11%±0.02% and 0.09%±0.01% reduction in A1c, respectively (p<0.0001), with a relative contribution of PPG to FPG of 82% On Lispro, a 1 mmol/L reduction in FPG and PPG was associated with reduction in HbA1c of 0.12%±0.02% and 0.07%±0.02%, respectively (p<0.0001), with a relative contribution from PPG to FPG of 58% On URLi, a 1 mmol/L reduction in FPG and PPG was associated with increase in BGM-derived Time in Range of 10% and 9%, respectively (p<0.0001) On Lispro, a 1 mmol/L reduction in FPG and PPG, was associated with an increase Time in Range of 10% and 9%, respectively (p<0.0001)
Time In Range In Children With Type 1 Diabetes In A Pediatric Unit	Retrospective study of 107 pediatric patients with type 1 diabetes on insulin pump therapy (mean age 13 years, 45% female, mean diabetes duration five years, mean insulin pump therapy duration three years) Time in Range was 49%, Time Above Range was 43%, Time Below Range was 5%; average 11 hypoglycemia events/day Total daily dose of insulin per kilogram (TDD/Kg) was 0.9 (0.3-1.7) units, 35±9% of daily basal dose, and 65±9% of daily bolus dose Diabetes duration was weakly correlated with Time Below Range (r=0.2), hypoglycemic events (r=0.2), and A1C (r=0.3); TDD/Kg was weakly correlated with TIR (r= -0.3)
Awareness Of Time In Range – Opportunities For Increased Adoption	Online survey of 985 adults with diabetes and 44 caregivers of children and/or adults with diabetes to assess awareness of Time in Range, its value and ease of use, and barriers associated with using Time in Range 44% of respondents were aware of Time in Range. Awareness was higher among people with T1D (67%, n=201) and T2D on MDI/pump therapy (58%, n=195) compared to other people with T2D (33%, n=606) 75% of CGM non-users (n=680) were unaware of Time in Range compared to just 14% of CGM users (n=322) Of those aware of Time in Range (n=445), most believed Time in Range was valuable (88%) and easy to use (89%) Among Time in Range non-users (n=559), 89% believed Time in Range would be helpful in diabetes management, and 66% identified education from their provider as a helpful resource for learning more about Time in Range
Confidence Intervals Estimation Of Predicted Hba1c Derived From Time-In-Range For Linear Regression Analysis	Outpatient study of 101 type 2s who underwent A1c testing, wore a FreeStyle Libre Pro, and did not change diabetes therapies on a hospital visit Two patterns of 32 patients were selected, each comprising eight patients at each of the following A1c levels: 6%, 7%, 8%, and 9% Pattern 1 included patients with low A1c and low Time in Range, among whom the ratio of Time Below Range (<70 mg/dL) to Time Above Range (>180 mg/dL) was negatively correlated with A1c Pattern 2 included patients in whom the Time Below Range:Time Above Range ratio was not correlated with A1c In Pattern 1, A1c was distributed normally while Time in Range was not. The center of curves for 95% confidence intervals estimation for predicted A1c derived from Time in Range was situated on the lower Time in Range side of the center of the Time in Range distribution range In Pattern 2, both A1c and Time in Range were distributed normally. The center of curves for 95% confidence intervals estimation for predicted A1c derived from Time in Range was situated at the center of the Time in Range distribution range
Predicting Time-In-Range Based On Past Glucose Responses From Similar Meals	Analysis of anonymized meal data from a subset of 100 SNAQ users who had recorded 10+ meals in the app and who had imported >33,000 glucose data points Time in Range predictions were calculated for 963 meals; on average, 5.6 similar meals were considered to predict Time in Range The calculated mean absolute error (MAE) was 23% and mean squared error (MSE) was 9% for Time in Range predictions based on past similar meals (n=578) This data suggests that predictions based on past responses can be a viable solution in predicting Time in Range if further information is incorporated, such as insulin or activity data
Nocturnal Glucose Fluctuations In Patients With Type 1 Diabetes: Which Patterns Are Associated With Hypoglycemia?	Retrospective clustering analysis of CGM records from 405 adults with T1D on basal bolus insulin therapy (median age 36, median diabetes duration 16 years, 64% female, and median A1c 8.1%) 14 clusters without hypoglycemia and nine clusters with nocturnal hypoglycemia (NH) were identified In only seven cases, NH was observed at the beginning of the nocturnal interval (0-1 a.m.). Mostly, NH was observed at 2–4 a.m. in clusters with initially normal glucose and downtrend. If glucose was initially elevated, NH was recorded more frequently at 4–6 a.m. The rate and amplitude of the rise in glucose levels after hypoglycemia varied significantly between clusters, affecting glucose levels at the end of the night These results suggest that clustering of nocturnal glucose dynamics is a promising approach for identification of T1D subjects at high risk of NH
Structured SMBG As a Measure Of Glycemic Variability In Persons With Type 2 Diabetes	Retrospective observational study to compare BGM values in/outside range with CGM values of Time in Range, Time Above Range, and Time Below Range in 12 type 2s Across 93 days of data, average BG calculated by BGM was 171 mg/dL and had strong correlation with average BG calculated by CGM (149 mg/dL) ( r=0.81, p<.001) Time in Range and Time Above Range showed strong correlation with BGM points in and above range (PIR and PAR), which were statistically significant: BGM(70-140) PIR vs. CGM(70-140) TIR: r=0.61, p<0.001 BGM (70-180) PIR vs. CGM(70-180) TIR: r=0.68, p<0.001 BGM (70-140) PAR vs. CGM(70-140) TAR: r=0.61, p<0.001 BGM (70-180) PAR vs. CGM(70-180) TAR: r=0.73, p<0.001)
Personalized Glucose-HbA1c Relationship For Clinical Management Of Individuals With Diabetes	Retrospective analysis of three months of CGM and A1c data from 216 type 1s (49% black, 55% female, mean age 30) to compare apparent glycation ratio (AGR) across groups Overall calculated KM value (glucose affinity for GLUT1) was 464 mg/dL with AGR showing differences in the white and black populations at 70 and 74 ml/g, respectively (p<0.001) AGR was highest in those aged >50 years at 75 ml/g, decreasing to 73 ml/g in 19-50 years, with a further drop to 71 ml/g in the youngest group (p<0.05) In contrast, AGR values were similar in men and women at 72 and 73 ml/g (p=0.27)
Estimated A1c Is Highly Correlated With Home Kit A1c In A Large Remote Diabetes Monitoring Program	Cohort analysis of 7,665 people with diabetes (mean age 67, 48% female, 75% white, 95% with T2D) Average self-reported baseline A1c was 7.1% and average home kit A1c was 6.9% Members had a mean change in home kit A1c of 0.13% (SD 0.70) and BGM-estimated A1c of 0.01% (0.64%). A strong, positive correlation was shown between BGM-estimated A1c and home kit A1c (r2= 0.69) Though there is positive correlation between home kit A1c and estimated A1c via BGM, estimated A1c via BGM slightly underestimates A1c increases at low home kit A1c values and overestimates A1c reductions at high home kit A1c values

New therapies in diabetes

Title	Details + Takeaways
Dual GIP-GLP-1 Receptor Agonist Tirzepatide Improves Glucose Control And Insulin Sensitivity In Mixed Meal Tests In People With Type 2 Diabetes (Mechanism Of Action Study)	28-week phase 1 study of tirzepatide 15 mg (n=45) vs. semaglutide 1 mg (n=44) vs. placebo (n=28) in people with T2D Tirzepatide reduced A1c by -2.05% vs. -1.64% with semaglutide vs. +0.29% with placebo (baseline 7.8%) Tirzepatide reduced body weight by -11.2 kg vs. -6.9 kg with semaglutide vs. 0 kg with placebo (baseline 94.5 kg) Tirzepatide improved insulin sensitivity as measured by M-value compared to semaglutide and placebo (8.98 vs. 7.45 vs. 5.42, respectively)
Type 1 Diabetes: Impact of SGLT-2 Inhibitors On Glycemic Control	Three-month retrospective study investigating dapagliflozin (5 mg) efficacy in people with T1D (n=17) After three months on SGLT-2, spent +2.2 hours/day in Range (p=0.019) with 60% Time in Range; coefficient of variation reduced three points to 40.7 (p=0.001); no significant difference on Time Above or Below Range After three months on SGLT-2, mean weight reduced 4.4 kg from 88.7 kg (p<0.001) and mean BMI reduced 1.5 kg/m2 to 29.3 (p<0.001)
Assessment Of Patient Satisfaction And Clinical Efficacy With Semaglutide In Suboptimally Controlled Patients With T2DM: A Study Based On The Flash Glucose Monitoring System	Three-month prospective study evaluating glycemic control and PROs in people with T2D (n=52) who were treated with liraglutide at baseline and switched to once-weekly semaglutide Three months on semaglutide significantly reduced A1c (7.8% vs. 8.1% at baseline; p<0.001) and body weight (84.6 kg vs. 87.2 kg at baseline; p<0.001) After three months on semaglutide, spent +3.2 additional hours/day in Range with 51% Time in Range and spent -2.5 fewer hours/day above range with 20% Time Above Range (all p-value < 0.001) All patients preferred/strongly preferred once-weekly semaglutide over liraglutide; comparable treatment satisfaction between males and females
The Cardiovascular Benefits of GLP1-RA Are Related To Their Positive Effect On Glycemic Control: A Meta-Regression Analysis	A meta-analysis of nine studies (n=64,236) investigating the association between A1c, weight, and SBP with GLP-1 CV benefits Found significant heterogeneity in GLP-1 MACE reduction across trials (p=0.03) Regression model found that a 1% reduction in A1c with GLP-1 was associated with a 21.9% reduction in MACE (p<0.001), indicating MACE benefits of GLP-1s are dependent on A1c reduction
DYRK1A Inhibitors In The Treatment Of Diabetes	Investigated the use of DYRK1A kinase inhibitors to promote glucose-stimulated insulin secretion in MIN6 cells DYRK1A kinase inhibitors led to increase C-peptide and endogenous insulin levels, while reducing glucagon levels DYRK1A kinase inhibitors led to signals indicating increased beta-cell proliferation and engraftment
Factors Associated With Different Patterns Of Weight Change After Bariatric Surgery: A Longitudinal Study	Longitudinal study of patients with laparoscopic Roux-en-Y gastric bypass or sleeve gastrectomy (n=196, baseline BMI 42 kg/m2) assessing factors associated with greater % weight loss 32% total weight loss evidence within one year after procedure, with weight stabilization after that period Nutritionist follow-up after surgery was associated with increased  total weight loss (+2.4%, p=0.014)
Dose Optimization Study (Dos): Simplified 2x Dose Of Inhaled Technosphere Insulin Provides Significant Reduction In Post Prandial Glucose Excursions With No New Safety Concerns	Dose optimization study (n=20 with T1D or T2D on basal-bolus insulin) comparing efficacy and safety of two pre-prandial technosphere insulin (Afrezza) doses: current US labeling vs. simplified higher dose Compared to current US labeling dose, the simplified higher dose provided significant reductions in post-prandial glucose excursions over 120 minutes after the meal (89 vs. 37 mg/dL, p<0.001) The simplified higher dose significantly lowered peak glucose levels from 229 mg/dL with the current US labeling dose to 179 mg/dL (p<0.001) There were no new safety concerns and no severe hypoglycemia with either dose
Use Of Basal Insulin-Glp1 Combination For Therapeutic Simplification In Type 2 Diabetic Patients With A Basal-Bolus Or Basal Plus Scheme	Observation study of patients on basal-bolus insulin (n=28) from January 2019 to July 2021, assessing efficacy of basal insulin-GLP-1 as simplification therapy Switch to insulin-GLP-1 led to significant A1c decrease of 0.8% (p=0.011) from baseline A1c of 8%, and led to significant -2.1 kg weight loss (p=0.004) from baseline weight of 76 kg Switch to insulin-GLP-1 led to significant insulin dose reduction from 51 units/day to 23 units/day (p=0.032)
Duodenal-Jejunal Bypass Liner For Treatment Of Type 2 Diabetes And Obesity: Four Year Outcomes In The First National Health Service (NHS) Endobarrier Service	Observation study assessing long-term benefits of EndoBarrier (a duodenal-jejunal bypass liner implanted for one-year) for refractory diabesity after EndoBarrier removal; From 2014 to 2017, implanted 62 EndoBarriers, all of which were removed by 2018 During EndoBarrier treatment, A1c fell 1.8% from 9.1%, weight fell 17 kg from 122 kg, systolic BP fell from 14 mmHg from 139 mmHg (all p<0.001) Three years post-EndoBarrier, 74% of participants (31/42 participants) maintained most improved achieved with EndoBarrier, while the other 26% reverted to baseline
Assessing Time In Range (TIR) With Postprandial Glucose (PPG)-Focused Titration Of Ultra Rapid Lispro (URLI) In Patients (Pts) With Type 1 Diabetes (T1D)	Phase 2, single-group, open-label study of 31 type 1s on MDI (n=31) using InPen bolus calculator and Dexcom G6 CGM, assessing TIR on ultra-rapid lispro compared to lispro during 11-day lead-in period Mean Time in Range during ultra-rapid lispro treatment was 70%, not significantly different from 11-day lead-in period with lispro Ultra-rapid lispro was associated with a significant -0.36% A1c reduction from baseline of 7.11% (p<0.001) Ultra-rapid lispro significantly reduced post-prandial glucose excursions by 16.6 mg/dL from 72 mg/dL (p<0.05) There were no differences in severe hypoglycemia or serious adverse events between lispro lead-in and ultra-rapid lispro
Efficacy And Safety Of Liraglutide In An Indian Adolescent Population With T2DM And Obesity: A Single Centre Experience From Eastern India	Prospective real-world observational trial (n=31, ages 12-17 years old) of Indian adolescents with type 2 diabetes and obesity therapy-naïve or on metformin, assessing efficacy and safety of liraglutide Liraglutide led to a significant A1c reduction of 2% from 8.1% (p<0.001) Liraglutide led to a significant body weight reduction of 4 kg from 69 kg (p<0.001) The most common adverse events were gastrointestinal in nature, with no severe adverse events or severe hypoglycemic events
Real-World Effectiveness Of iGlarLixi Therapy In Outpatients With Type 2 Diabetes: The Solo Retrospective Cohort Study	Real-world retrospective cohort study of adult type 2s (n=383) assessing efficacy of iGlarLixi; at baseline, 65% were on oral antidiabetic drugs, and the rest were on basal insulin plus oral antidiabetic drugs iGlarLixi led to a significant A1c reduction of 1.4% and 1.7% after six and 12 months, respectively, from baseline of 9.14% (p<0.001) iGlarLixi led to a significant reduction in body weight by 2 kg and 3 kg at six and 12 months, respectively, from baseline of 102 kg (p<0.001)
Early Deescalation With Ideglira In Patients With Type 2 Diabetes Using Short-Term Human Basal-Bolus Therapy To Correct Severe Hyperglycemia	Prospective, 12-month, real-world, single-arm trial assessing the use of iDegLira for deescalating short-term basal-bolus therapy for correcting severe hyperglycemia in people with type 2 diabetes (n=44) After 4 months on iDegLira, A1c significantly decreased from 12.2% to 6.1%  (p<0.0001) iDegLira led to a nearly 50% decrease in mean total daily insulin dose from 41 IU
Experiences Of Caregivers Of Children And Young Adults With Type 1 Diabetes Related To Severe Hypoglycemia And Being Prepared With Nasal Glucagon– A Qualitative Study	Cross-sectional, qualitative study investigating experiences of caregivers (n=32) for children and young adults with type 2 diabetes Theme of proactivity among caregivers to prevent severe hypoglycemia, citing acute distress due to hypoglycemia, social inhibition, and sleep loss due to hypoglycemia prevention and treatment Many caretakers described nasal glucagon as increasing their confidence in being prepared for severe hypoglycemia
Glucagon Awareness And Utilization In Children With Type 1 Diabetes	Qualitative study evaluating experiences of children with type 1 diabetes and caregivers (n=34) of using glucagon emergency kits 88% of patients (30/34 patients) had at least one glucagon rescue medication, with four having used it in the past 26% of patients (nine families) realized the medication had an expired shelf life 24% of patients (eight families) expressed a need to review how to use glucagon
Population Pharmacokinetic Modeling Of Dasiglucagon In Subjects With Type I Diabetes Mellitus	Pharmacokinetic study of dasiglucagon in people with type 1 diabetes (n=337) PK model predicted lower dasiglucagon exposure following injection in thigh, buttocks, or deltoid vs. abdomen, and predicted lower exposure in pediatric vs. adult patients PK model predicted higher exposure with lower eGFR and faster absorption in female patients
Inhaled Glucagon, A New Well-Accepted Therapeutic Tool In Pediatrics	Descriptive study evaluating clinical experience in initiating Baqsimi (nasal glucagon) in children with type 1 diabetes (n=68) who previously used Glucagen HypoKit. All participants demonstrated correct use of Baqsimi Three episodes of severe hypoglycemia, all of which resulted in Baqsimi use and almost immediate recovery without sequelae
Residual Β-Cell Function In Subjects With Type 2 Diabetes Protects From Non-Adherence To Insulin Therapy Versus Subjects With Type 1 Diabetes	Computer simulation study investigating the effect of inconsistent insulin therapy use in type 1 and type 2 diabetes over six months Computer simulation found that delaying/ skipping pre-meal insulin bolus led to a significantly reduction Time in Range by -3.6 hours/day in people with type 1 diabetes Computer simulation found that delaying/ skipping pre-meal insulin bolus led to a significantly reduction Time in Range by reduced Time in Range by 34 minutes/day in people with type 2 diabetes

Human factors, big picture and other

Title	Details + Takeaways
Impact Of New Technologies On Quality Of Life And Glucose Control In Patients With Type 1 Diabetes	Evaluating different devices on glucose control and quality of life outcomes in 69 patients with type 1 diabetes 36 patients on MDI and 33 patients on CSII were evaluated using the Diabetes Treatment Satisfaction Questionnaire (DTSQ), the Diabetes Specific Quality of Life Scale (DSQOLS), and the Short Form Health Survey (SF-36), as well as on A1c and TIR metrics Patients in CSII group had higher treatment-related satisfaction (84.8% vs 52.8%, p=0.005) and better disease acceptance (84.8% vs 52.8%, p=0.012); patients with CSII also had better TIR (p=0.001) Participants on Dexcom G6 had higher TIR than participants on Freestyle Libre (p=0.03) and had similar TIR values to participants on Medtronic 640G and 670G (p=0.12); there were no differences patient reported outcomes between different pumps Overall, pumps may improve quality of life over MDI treatment; may also be differences between glucose monitors that can impact glucose control
Glycaemic Control in Patients Using Different Insulin Delivery and Glucose Sensoring Devices: Real-Life Data from Tampere University Hospital	Evaluated the glycemic control of patients (n=1,464, 79% with T1D) based on their insulin delivery and glucose measuring outcomes Poor glycemic control was found to be independently associated with young age group and high daily insulin dose per weight Hybrid closed loop or sensor augmented pumps yielded the best results; intermittent CGM with MDI was independently associated with poor glycemic control
Comparison of Attitudes of Physicians, Parents and People with Diabetes Towards Digitalization and New Technologies in Diabetes	Evaluating differences in attitudes of physicians, parents, and people with diabetes regarding digitalization and new technologies Participants surveyed in 2019 (n=324 diabetologists; n=3,427 PwD) and again in 2021 (n=305 diabetologists; n=2,417 PwD) to understand how perceptions have changed over time Parents of children with type 1 diabetes had more positive attitudes toward digitalization than patients with type 1 diabetes and diabetologists The perceptions of digitalization improved across all four groups in the two-year period PwD (type 1 and type 2) have significantly better perceptions of digitalization and tech than physicians AID systems were ranked as the number one priority for future of endocrinology, followed by interoperability and software for analyzing glucose data and AI – consistent across all groups
Exploring The Understanding And Confidence Of Inpatient Management Of Insulin Pumps And Glucose Monitors Amongst Healthcare Professionals	Evaluated the understanding and management of insulin pumps and glucose monitors among HCPs 10-question survey distributed to doctors and nurses at the Barnsley Hospital NHS Foundation Trust assessing awareness of and confidence in managing diabetes devices; 31 of 439 eligible surveys were completed 71% of respondents had cared for a patient on insulin pump therapy; 29% of them did not feel confident in their management; 74% of participants had heard of glucose monitoring and 58% correctly identified rapid-acting insulin as the infusion for pumps Varying levels of experience, knowledge, and confidence among HCPs speaks to the need for additional training and educational events to improve patient care
How Do Physicians Rate the Indication for Modern Technologies In People With Diabetes?	Evaluating how different diabetologists (n=305) consider indications for modern technologies with diabetes Diabetologists ranked children and adolescents as the highest priority for devices, then adults with T1D, pregnant people with diabetes, and adults with T2D on intensified insulin therapy; generally less support for people with T2D on conventional insulin therapy Diabetologists see a broad indication for diabetes technology, independently of the reimbursement situation
Ranking Of Important Factors In Recommending Diabetes Technology By Pediatric And Adult Endocrinologists: Data From The T1D Exchange Health Equity Advancement Lab (T1DX Heal) Study	Comparing rates of CGM and pump usage between non-Hispanic black, Hispanic, and non-Hispanic white patients and understanding prescribing preferences of providers Used survey data from seven centers participating in the T1DX-QI study (n=35 adult providers, n=75 pediatric providers) Pediatric providers ranked patient technology preference, self-management of blood glucose, and A1c as the most important factors in recommending technology while adult providers ranked patient preferences, insurance type, and self-management of blood glucose as the most important factors No differences in recommendation factors were identified based on race; however, there were some different considerations depending on the type of technology (pump vs CGM) Conclude that understanding the factors that providers consider when prescribing technology can help develop appropriate solutions to target technology inequities
Race-Ethnicity Mediated Bias In Recommending Diabetes Technology: Does Implicit Bias Training Make A Difference?	Evaluating the role of diabetes provider implicit bias mediated by patient’s race in recommending diabetes tech to patients Providers across seven T1DX-QI centers (n=35 adult providers, n=75 pediatric providers) completed a Diabetes-Provider Implicit Bias tool containing clinical vignettes and ranking exercises Implicit bias was identified in 34% of providers in the survey; no differences in age, provider role, provider type, practice setting, or number of practice years were identified between the biased and non-biased groups 89% of providers in the bias group believed that they were able to recognize their own bias compared to 61% in the non-biased group; the bias group also had a higher rate of people reporting previous bias education and training
Age-Based Disparities In Patient-Provider Discussion Of Diabetes Technology: A Call For Dismantling Ageism In Diabetes Care	Evaluating age-based differences in patient reports of HCP-initiated discussions of CGM and insulin pump technology PWD (n=5,435) surveyed on provider-initiated conversations with their HCPs about CGM and pump technology Older age was a predictor of lower likelihood of technology discussion; 8% of non-CGM users (age 65+) reported a discussion about CGM compared to 53% for ages 18-35 and 81% for ages 36-64 A similar pattern was observed in patients on MDI: 31% of patients age 65+ reported a conversation about pumps compared to 81% for ages 18-35 and 51% for age 36-64 Results were consistent even after controlling for A1c, income, insurance status, race, HCP type, and diabetes type; trends were observed in both patients with type 1 and type 2 diabetes
The Impact Of Socioeconomic Deprivation On Access To Diabetes Technology In Adults With Type 1 Diabetes	Evaluating the relationship between area deprivation and socioeconomic status with access to diabetes technology and the subsequent outcomes in adults with T1D Retrospective analysis from three hospitals in the UK (n=1,631) Technology use was highest in the least deprived quartile and lowest in the most deprived (67% vs 44%; p<0.0001) but there was no association between deprivation and A1c-related outcomes of technology (p=NS); white adults were more likely to use technology than Black adults (59% vs 40%; p<0.05)
Income- and Insurance Status-Based Disparities in Patient-Provider Discussion of Diabetes Technology Among People With Type 1 Diabetes	Evaluating income- and insurance- based differences in T1D patient reports of provider-initiated discussions about CGM and pump tech Participants (n=2,283) filled out an online survey asking about HCP-initiated conversations about CGM and pumps Non-CGM users with income <$50,000 were less likely to report a conversation about CGM (36%) than non-CGM users with incomes of $50,000-$100,000 (58%) and > $100,000 (73%) Those with private insurance were more likely to discuss CGM (63%) than those on Medicare (38%) Similar patterns were observed with pumps, with highest income patients reporting highest rates of conversation with their providers Results suggest that income- and insurance- based biases may impact HCP’s likelihood to initiate conversations about diabetes technology
Assessing Expectations Of Hybrid Closed Loop (HCL) Insulin Delivery Among Underserved Youth With Suboptimally Controlled Type 1 Diabetes (T1D) And Their Caregivers	Evaluating expectations related to AID among underserved youth with type 1 diabetes and their caregivers 30 publicly insured non-Hispanic black youth with A1c>10% filled out INSPIRE questionnaire to measure expectations of AID tech (higher results reflect greater positive expectations) The positive expectancy of AID systems was comparable between underserved youth in study and the predominantly non-Hispanic white, privately insured, high-income cohort in the Tandem Control-IQ pivotal trial Results suggest that different perceptions in hybrid closed loop technology do not explain disparities in T1D technology usage
Tech Truths: Reflections on Technology Utilization from Youth and Their Parents After T1D Diagnosis	Evaluating families perceptions of technology use during the first year post diagnosis Participants (n=21 patients, n=39 parents) were interviewed via videoconference to examine the challenges and successes of technology utilization Participants reported improved ability to monitor glucose levels, decreased worry about out-of-range numbers, and some occasional conflict about site and notifications; challenges primarily related to annoyance of alarms and questions and comments from peers Parents reported higher rates of optimism, improved quality of life, improved sleep quality, and increased comfort of child to be away from home Teenagers reported increased independence and greater opportunities for flexibility with food on pump therapy
Reasons For Hesitancy Toward New Automated Insulin Delivery System Adoption Among Adults Living With Diabetes In The United States, Canada, And Europe	Evaluation of the reasons for PwD’s unwillingness to adopt new AID systems Participants (n=5,226) from eight countries took an online survey asking about preferred AID systems after viewing the profiles of three new or upcoming commercial products; 975 respondents said they would not choose any of the products and the subsequent 808 qualitative write-in responses were coded and analyzed The top reasons for AID system adoption reluctance were: (i) unwillingness to switch therapies (19%); (ii) unwillingness to use an insulin pump (16%); and (iii) distrust of technology (12%) Respondents in the US were more distrustful of technology (15%) than those in Canada and Europe (7%) Conclude that reducing hesitancy toward AID system adoption would likely involve reducing insulin pump intrusiveness, increasing patient knowledge, and increasing system customizability
Human Factors Influencing Frequency and Accessibility of Blood Glucose Self-Monitoring Among Adults with Type 2 Diabetes	Evaluating potential barriers to diabetes technology access among infrequent BGM users and non-BGM, non-CGM users with T2D Responders to an online survey (n=798) were asked about factors affecting their BGM frequency and factors that could increase it Infrequent BGM participants cited forgetfulness (43%), pain with fingersticks (14%), and challenges with affordability (14%) as primary barriers; non-BGM, non-CGM users said they didn’t know enough about diabetes technology (18%) and felt intimidated by its use (14%) Respondents in both groups said that improving affordability and greater encouragement from HCPs may also increase uptake
Daily Predictors of Diabetes Self-Management in Adolescents With Type 1 Diabetes (T1D) Using CGM	Evaluating the daily factors that influence adolescents’ and young adults’ engagement in diabetes self-management Participants (n=100) took part in a two-week prospective study where they chose a management goal to focus on and were surveyed on six random days with a 26-question predictor of current glucose level, sleep, mood, motivation, control beliefs, social support, illness, self-esteem, and help needs Mood, social support, and control beliefs were found to be predictive of diabetes management engagement
Eversense As the First Choice Of CGM To Improve Quality Of Life Among Patients With Type 1 Diabetes	Evaluation of patient satisfaction, health-related quality of life, and fear of hypoglycemia in patients with Eversense CGM as their first CGM Participants (n=10) on Eversense were evaluated three months after implementation Patients showed no significant fear of hypoglycemia and positive satisfaction for treatment even when Eversense was used as the first choice CGM
European Survey on Adult People With Type 1 Diabetes (T1D) and Their Caregivers: Insights Into Use of Technological Devices and Digital Tools	Evaluating impact of diabetes on daily life, including use and perception of technology in T1D management Respondents (n=458) filled out an online survey asking about technology usage and familiarity with devices Smart insulin pens were more common in patients ≤40 years, classic glucose meters used by 43% of PwD; 64% used continuous or flash glucose monitoring; half of participants felt they lacked knowledge about T1D but most believed that digital tools were useful Authors conclude that people with T1D and their caregivers require more support and information on T1D management and that the majority of PwD believe that devices and tech are/could be useful
Current Insulin Infusion Set Criteria Do Not Represent Real-Life Setting and May Skew Infusion Set Failure Outcomes In Extended-Wear Infusion Set Studies	Goal is to challenge currently accepted insulin infusion set failure criteria which are too stringent for home-use extended-wear infusion set studies and are more suitable for controlled inpatient settings within a research facility Insulin infusion set data was retrospectively analyzed from 13 participants who wore a total of 66 infusion sets during 52 wear periods of up to 14 days Nearly 70% of infusion sets were kept in place between 0.2 and 11.9 days longer (median 3.2 days) than stated in protocol set failure. Remaining infusion endpoints were removed at the stated time Conclude that PwD have good knowledge of their glucose profiles and will choose to leave an infusion set in place longer than predefined in the study protocol or recommended by manufacturers’ instructions Recommend less stringent failure criteria that more closely mimics the real-world decision processes for future use in extended wear infusion set studies
Green Diabetes: Packaging Waste and Sustainability	Evaluating the ecological desires of PwD to determine how they are thinking about the waste associated with diabetes Participants (n=2,417) surveyed about their attitudes and assessments of AID systems Found that while most participants said they would like to see more reusable devices in diabetes therapy (67%); however, the waste associated with a product is only a deciding factor when selecting devices for 15% of PwD Authors conclude that despite the stated preference for reduction of waste, there is still a gap between the attitude and actual behavior
Durability of 10-Khz Spinal Cord Stimulation for Painful Diabetic Neuropathy: 18-Month Multicenter Randomized Controlled Trial Results	Evaluation of 10-Khz spinal cord stimulation (Nevro Corp) on patients with painful diabetic neuropathy Participants (n=216) were randomized 1:1 between 10-kHz SCS treatment and conventional management Participants in the SCS group experienced substantial, sustained pain relief over 18 months No stimulation-related neurological deficits and no explants due to ineffectiveness over the course of the study were reported; five explants occurred due to procedure-related infections and one occurred as a precaution for endocarditis

COVID-19 and Diabetes

Title	Details + Takeaways
European Survey On Adult People With Type 1 Diabetes (T1D): Insights Into The Impact Of COVID-19	Online survey of 458 adults with T1D and 54 caregivers to evaluate the impact of the COVID-19 pandemic on T1D management The majority of participants (71%) worried about the pandemic, despite only 8% reporting being significantly affected by COVID-19-related illness 58% half of PWD and 52% of caregivers reported no difficulty controlling T1D because of the pandemic, 30% of PWD and 39% of caregivers reported a little difficulty, and a 12% of PWD and 7% of caregivers reported a lot of difficulty The pandemic changed T1D management for two-thirds of PWD. 33% of participants reported cancelling/postponing some medical appointments, 22% reported more frequent telehealth use, 15% reported having more time to care for themselves, and 11% reported more regular and balanced meals
Remote And In Clinic Initiation Of Advanced Hybrid Closed Loop System Minimed 780G In Children And Adolescents With Type 1 Diabetes	The study evaluated glycemic control between remote and in-clinic initiation of MiniMed 780G hybrid closed loop system in 64 youth with T1D, ages 7 to 18 Both groups saw comparable declines in A1c, with A1c in the remote group decreasing from 8.3% at baseline to 6.7% at the end of the study (p=0.002) and A1c in the in-clinic group decreasing from 8.2% to 6.2% (p=0.001) Both groups saw similar increases in Time in Range, with the remote group increasing from 49% at baseline to 74% at the end of the study and the in-clinic group increasing from 48% at baseline to 78% at the end of the study Based on these findings that A1c levels, TIR, and SmartGuard use did not differ significantly by group, the researchers suggest that remote initiation should be offered as an alternative to in-clinic initiation
Impact Of Finances On Diabetes Care And Supply Access During Covid-19 Among U.S. Adults Living With Diabetes	Among 4,991 adult respondents with diabetes, 24% reported that finances had a significant influence on diabetes care 13% of respondents delayed care due to cost, with respondents in the lower income bracket (<$50k) more likely to delay care compared to those in the middle ($50k-$100k) and higher (>$100k) income brackets Among respondents who delayed care due to cost, 50% of those on MDI delayed a refill of insulin and 50% of pump users delayed an order of pump supplies. Among CGM users who delayed care, 48% delayed an order of CGM supplies
Lessons Learnt From COVID-19 For Health Systems: The Use Case Of Diabetes Remote Monitoring	In depth needs assessment among multidisciplinary experts across Europe, with roundtable discussion in September 2021 All experts agreed that incorporating telehealth as a standard of care in diabetes faces significant challenges, including: (i) a fragmented approach to healthcare technology assessment and reimbursement; (ii) lack of eHealth education and literacy among HCPs and patients; (iii) lack of data integration with EHRs; and (iv) patient consent, privacy, and data protection The experts identified key actions to address these challenges: (i) investment in telemedicine and technology that allows remote monitoring; (ii) bridge health inequities; (iii) protect patient data and privacy; (iv) incentivize broad adoption by patients, providers, and health systems; and (v) ensure a true measure of the value these technologies bring
Evaluation Of Glycaemic Control In Patients With Diabetes Mellitus Hospitalized Due To COVID-Associated Pneumonia	This study aimed to compare glycemic control and therapeutic changes during the inpatient stay in patients with an A1c <7% vs. those with A1c ≥ 7.0% on admission in 59 patients hospitalized in late 2020 due to COVID-19 illness Patients with higher baseline A1c consistently showed higher glucose values during their stay (299 vs. 205 mg/dL, p<0.001) Insulin requirements increased in both groups, with a significantly higher maximum insulin dose of 52 IU per day in those with baseline A1c ≥7.0% vs. 13 IU per day in those with baseline A1c <7.0% Mortality was similar across the groups, with patient’s age as a significant risk factor for hospital mortality
Sars-Cov-2 And Type 1 Diabetes In Children: Is There A Difference In Incidence Or Patient Characteristics During The Pandemic? Preliminary Results	Observational single center study of children (<16 years) with newly diagnosed T1D in Antwerp, Belgium that compared those diagnosed between January 2015 – February 2020 (n=203) with those diagnosed from March 2020 to February 2021 (n=51) No significant different in incidence or patient characteristics during the two years of pandemic compared to previous years (age, A1c, C-peptides, and ketone bodies) No delay in seeking medical care and no difference in severity of disease at presentation; no higher need for insulin at discharge during the pandemic vs. previous years
Effect Of The COVID-19 Pandemic On Metabolic Control In Insulin Pump Users In A Diabetes Center In Colombia	Observational study to compare metabolic management prior to and during the pandemic following implementation of a telemedicine program Examined 44 insulin pump users from January 2020 to July 2021; patients were 50% male with a mean age of 41 years and diabetes duration 21 years There were no significant changes in metabolic control or Time in Range. Mean A1c was 7.7% pre-pandemic and 7.5% post-pandemic, while Time in Range was 74% pre-pandemic and 74% post-pandemic Suggests that the telemedicine program for follow-up of patients on insulin pumps supports maintenance of metabolic management
Impact Of COVID-19 Lockdown On Glycemic Control In Patients With Type 1 Diabetes. What Has Happened 1 Year After Lockdown?	Observational retrospective study of FreeStyle Libre data collected from type 1s during the 2 weeks prior to lockdown start, the last 2 weeks after 8 weeks of consecutive lockdown, and the last 2 weeks after one year of lockdown in Spain Data analyzed from 287 patients with median age 46 years, 50% male, and median diabetes duration 21 years. Median lockdown time was 54 days. After one year of lockdown, there was an improvement in A1c and Time in Range in 42% and 47% of participants, respectively. This was comparable to an improvement in A1c and Time in range in 48% and 47% of participants, respectively, after 8 weeks after lockdown The study found a “relative change” in both A1c and Time in Range without increased rates of hypoglycemia in 16% of participants one year after lockdown, compared to 10% of participants 8 weeks post-lockdown Overall, type 1s using FreeStyle Libre during the COVID-19 pandemic saw improvements in glycemic control, which were sustained one year after lockdown
Depression And Anxiety Among Patients With Type 2 Diabetes In COVID 19 Pandemic	Cross-sectional questionnaire-based study conducted among 80 adults with T2D in Tunisia to assess prevalence and factors associated with depression and anxiety during the COVID-19 outbreak The prevalence of depression and anxiety symptoms were 60% and 53%, respectively A1C ≥9%, high educational level, obesity, previous cardiovascular events, and low income were risk factors associated with depression and/or anxiety (p<0.05) Fear of COVID-19 infection and requiring hospitalization or death from COVID-19 were the most stressful factors reported by patients (p=0.01)
Evaluation Of Complications And Outcomes In Patients With Diabetes Mellitus And COVID-19	Retrospective analysis of 1,500 medical histories from hospitalized patients with a confirmed COVID-19 diagnosis in Kazakhstan (n=235 PWD); patients were 63% female with a median age of 63 Acute respiratory distress syndrome, acute respiratory failure, and oxygen insufflation requirement were more common in the presence of diabetes (all p<0.001) Artificial pulmonary ventilation and requirement for ICU treatment were 3.7 and 2.4 times higher among PWD, respectively (p <0.001). Mortality was 2.1 times higher than in the group of patients without diabetes (p <0.001) Primary causes of mortality included acute cardiopulmonary failure (28.2%), pulmonary embolism (24%), multiple organ failure (10.9%), infectious-toxic shock (30.4%), other types of shock (2.2%), cerebral edema (4.3%)
The Protective Role Of The FGM System In Identifying Type 1 Diabetic Individuals At Risk Of Acute Diabetes Complications During The Sars Cov2 Pandemic	Retrospective analysis investigating Flash Glucose Monitoring in 56 T1s on multi-injection therapy in Italy during three separate periods: Period 1 (October 2020), Period 2 (May 2021), and Period 3 (November 2021) The coefficient of variation (CV) and the GMI increased significantly between period 1 and 2 (37.17 vs 42.30, p=0.03 and 8.08 vs. 9.1, p=0.005 respectively) In period 2, all subjects were called back to the center to be re-evaluated. The opposite trend was observed for CV and GMI from the second period to date (current CV: 33.10, p=0.0002  and GMI: 7.8, p=0.0004) TIR, TAR, TBR were not significantly different between Period 1 and 2, but they were significantly different from Period 2 to date (TIR2 55% vs TIR3 66%, p=0.001 and TAR2 40% vs TAR3 28%, p=0.001)
The Effects Of The COVID-19 Pandemic On People With Type 2 Diabetes Using One Drop	Survey study of 171 One Drop users with T2D to describe self-reported impact of pandemic on life areas, preventive care appointments, and health management behaviors. Mean age 55; participants were 55% male, 70% white, and 95% insured 94% reported negative life impacts, 63% reported missed/delayed preventative appointments, and 76% reported impacts on diabetes-related health behaviors Missed/delayed appointments were dental cleanings (42%), A1C or blood pressure checks (35%), and primary care visits (25%) For health behaviors, physical activity (68%), vegetable intake (32%), and carb tracking (37%) were most frequently impacted For those reporting impact, higher levels of impact were associated with higher levels of diabetes distress and A1C in each impact area (p<0.05), however, missed appointments were unrelated to A1C
The Impact Of COVID-19 Pandemic In Patients With Type 2 Diabetes	Study to examine the impact of COVID-19 lockdown on glycemic control in 78 T2s in Tunisia (41% male) In the period of three months after lockdown ending, participants saw worsening diabetes control, with a 1.9 mmol/l increase in mean fasting glucose and 1.1% increase in A1c. After lockdown, 41% of patients weren’t meeting their glycemic targets, compared to 26% of patients before lockdown. Participants also saw weight gain (+3.2 kg) and an increase in abdominal circumference (+2.7 cm) 6 patients were hospitalized during this period for major hyperglycemia, with 4 developing ketoacidosis and 2 developing hyperglycemic hyperosmolar syndrome

2021 ATTD Yearbook

COVID-19 pandemic and diabetes care

• Dr. Satish Garg (Barbara Davis Center) kicked things off on Friday morning with his overview of telemedicine and virtual diabetes care during the COVID-19 pandemic. Dr. Garg reflected on the >500 million known cases and >6 million known deaths since the start of the pandemic in 2020 and stressed the importance of vaccinating everyone on the Earth. Until we do that, Dr. Garg said, “This thing is not going to go away. We appreciated Dr. Garg’s focus on the disparities that were exacerbated by the pandemic, as evidenced by the uneven distribution of Time in Range improvements following rt-CGM initiation in the US (Hirsch et al., 2021). Dr. Garg also referenced the “convincing data” used to support the initiation of AID system remotely during the pandemic (Beth Happ et al. 2020), a notion fortified by dQ&A data presented at EASD 2021. Looking ahead, Dr. Garg expressed his hope that reimbursement parity for telemedicine visits would continue into the future, and questioned whether COVID-19 vaccinations would begin to be required annually in the future.

Insulin delivery hardware: pumps and pens

• Dr. Rayhan Lal of Stanford presented on the insulin delivery technology section of this year’s ATTD Yearbook, which was renamed to “Insulin delivery hardware: pumps and pens.” This change reflects last year’s update to the insulin pump chapter to also include the growing body of literature around smart insulin pen technology. Dr. Lal highlighted a wide array of literature. Starting with pens, Dr. Lal presented data showing that the Insulclock smart pen cap drove significant A1c reductions in people with uncontrolled type 2 diabetes (Galindo et al. 2021). Turning to pumps, Dr. Lal highlighted a collection of papers studying type 1s, showing that: (i) pump programs are associated with fewer age-related glycemic disparities (Mooney et al. 2021); (ii) type 1 US youth from lower SES backgrounds showed improved glycemic control on pumps vs. MDI (McKee et al. 2021); (iii) decreased DKA, less, severe hypoglycemia, and lower A1cs after pump utilization regardless of age reduction (Jeyam et al. 2021); (iv) reduced retinopathy progression (Reid et al. 2021) from pumps; and (v) the power of biphasic boluses for complex meals (Metwally et al. 2021). On pumps in type 2s, Dr. Lal touched on: (i) reduced A1c, fewer self-reports of hypoglycemia, and reduced total daily dose (Carlson et al. 2021); (ii) “tens of thousands” of associated injuries potentially as a result of pumps in the FDA MAUDE database that “are not easily summarized” (Krouwer 2021); (iii) fewer serum or capillary glucose readings out of Range (Halstrom et al. 2021); (iv) greater delivery errors over shorter intervals and volumes (Ziegler et al. 2021); and (v) unique incorporations of insulin on board into various bolus calculators (Buchanan et al. 2021).

New insulins, biosimilars, and insulin therapy

• Dr. Thomas Danne (Hannover Medical School) discussed advancements in insulin therapy in the last year highlighting Novo Nordisk’s insulin degludec for which top line results from the phase 3 ONWARDs trial were posted yesterday. Dr. Danne emphasized the longer eight-day half-life of insulin icodec and discussed a publication on outcomes among patients with type 2 diabetes initiating insulin icodec with no prior insulin treatment that demonstrated improvements in glycemic management. Additionally, Dr. Danne raised questions regarding the dosing and titration, hypoglycemia risk, and ideal candidates for once-weekly insulin. Dr. Danne also discussed data from two other studies demonstrating improvements in Time in Range with faster insulin aspart and improvements in postprandial glucose management with ultra-rapid lispro via CSII pump therapy. Regarding the future of ultra-rapid insulin use in pump therapy, Dr. Danne raised questions related to how much algorithms can or should be adjusted to the specific types of insulin being used, which could certainly have an impact on insulin active time and thus calculations for insulin on board.

Decision support systems and closed-loop

• Dr. Boris Kovatchev (University of Virginia) highlighted seven publications on decision support and closed-loop systems. Starting with decision support, Dr. Kovatchev discussed the ADVICE4U and KNN-DSS trials demonstrating non-inferiority of AI recommendations compared to expert clinicians. Turning to AID systems, Dr. Kovatchev discussed the remarkable Time in Range improvements from the pediatric Control-IQ study, the FLAIR study with MiniMed 780G, and Insulet’s Omnipod 5 pivotal trial. Dr. Kovatchev also looked to the future showing results from a small study (n=23) of people with type 1 diabetes using an insulin + glucagon fully closed loop system who saw overall Time in Range improvements of over 30% – these results are especially impressive as participants did not make meal announcements, which has been a limiting factor in the ability of AID systems to achieve target Time in Range. Finally, Dr. Kovatchev expressed excitement about the results of the DAPADream trial evaluating the add-on effect of SGLT-2s to AID, which demonstrated strong Time in Range improvements and adds to a growing body of literature demonstrating the benefits of SGLT-2s in people with type 1 diabetes.

Using digital health technology to prevent and treat diabetes

• Glooko’s Dr. Mark Clements discussed the quickly evolving digital health arena, outlining several themes seen over the last year: (i) movement toward more rigorous studies evaluating digital tools; (ii) increasing integration of interventions into one cohesive platform; (iii) increasingly remote trial methodology; and (iv) payers increasingly embracing the value of and payment models for digital therapeutics and remote patient monitoring. He also highlighted three publications on evaluating three digital solutions, drawing attention to specific sub-findings of the study that offer insight into what makes digital tools successful therapeutic interventions. The first study, published in DT&T, evaluated the DIABEO app with or without support from trained nurses; however, what Dr. Clements found most interesting was the distribution of app engagement, which showed very limited app engagement, something that is common across digital solutions. Moving on, the second study that Dr. Clements presented offered real-world analysis of the Novo Nordisk app, which is powered by Glooko. Again, what Dr. Clements found most interesting was the data on engagement, which showed that engagement with each feature of the app fell over time. However, those tools and features that didn’t require manual insertion of data saw higher engagement to start and maintained a higher level of engagement over time. Dr. Clements used this to highlight the need for further automated data collection and integration. Finally, the third study that Dr. Clements highlighted was an RCT evaluating the relative glycemic improvement seen with One Drop app use vs. use of the One Drop app that has automatic uploading of fitness tracker data, which indicated that automated uploading significantly improved A1c, further illustrating the importance of automatic data integration to drive better outcomes through digital tools.

Technology and pregnancy

• Dr. Jennifer Yamamoto (University of Calgary) gave the very important Yearbook update on diabetes technology in pregnancy. Dr. Yamamoto highlighted two studies, both of which point to the massive need to better support pregnant people with diabetes in achieving their glycemic targets and improved maternal and fetal outcomes. The first study was the UK’s National Pregnancy and Diabetes Registry data, which was published in The Lancet, and showed that after adjusting for maternal characteristics, no clinic was doing well in supporting pregnant people achieve targets. The second was the LOIS-P study, which was published in DT&T. LOIS-P included 25 pregnant people with type 1 in the US who wore a CGM and pump, offering a look into the specific glycemic and insulin dose changes that occur over the course of pregnancy, enabling a better understanding of where further treatment adjustments and more attention is needed. Overall, only seven participants in the study achieved the Time in Range targets, once again showing the massive need to better support pregnant people with diabetes to achieve glycemic management targets. To close, Dr. Yamamoto briefly drew attention to six studies on AID in pregnancy that are currently underway (NCT04492566, NCT04520971, NCT03774186, NCT04938557, NCT04420728, NCT04902378), several of which are nearing completion of recruitment or have completed recruitment.

Diabetes technology and therapy in the pediatric age group

• Stanford’s Dr. David Maahs took on the challenging task of trying to recap the busy year in diabetes technology and therapy in pediatrics in just six minutes. First, Dr. Maahs highlighted two articles on hybrid closed loop systems in pediatrics: one with Tandem’s Control-IQ and one with Medtronic’s MiniMed 780G. Dr. Maahs recapped results from Tandem’s pivotal trial of Control-IQ in children ages 6-13 which showed Time in Range increased 3.4 hours/day with the use of Control-IQ, a significantly greater improvement than the sensor-augmented pump/Basal-IQ group. Dr. Maahs also pointed at the results from the FLAIR study, which was read out at ADA 2020 and published in The Lancet in January 2021 showing superiority of Medtronic’s MiniMed 780G system over MiniMed 670G in a traditionally challenging population of adolescents and young adults (ages 14-29) with type 1. For the back half of his presentation, Dr. Maahs homed in on data from several registries published in 2021. This included data from the SWEET registry demonstrating increasing technology use and lower A1c results for children and adolescents using CGM and those using insulin pumps with an additive benefit for those using CGM & pump.

Advances in exercise and nutrition as therapy in diabetes

• This year’s remarks from Dr. Michael Riddell (York University) around exercise and nutrition as therapy in diabetes focused on a meta-analysis of low carb diets in people with type 2 diabetes. The study, published in BMJ in 2021, looked at 23 total studies of low and very low carb diets vs. low fat diet controls. In total, the 23 studies included 1,357 participants with type 2 diabetes. At six months, low and very low carb diets were associated with higher rates of diabetes remission (defined as A1c <6.5%), at 57% vs. 31%. However, Dr. Riddell pointed at a couple caveats of the study, noting that there were no significant differences when using a more stringent remission definition of A1c <6.5% and no diabetes medications and that the studies also showed relatively poor adherence to low and very low carb diets beyond six months.

Primary care and diabetes technologies and treatments

• Drs. Thomas Martens and Greg Simonson (International Diabetes Center) provided this year’s update on diabetes technologies and treatments in primary care, stressing that they would focus on both technology and therapy given the huge role of the latter in primary care. Dr. Martens started by highlighting the real-world retrospective study from Kaiser Permanente finding that type 2s saw a 0.6% A1c drop after initiating real-time CGM compared to non-users. As a reminder, this study was published in JAMA last year close to ATTD 2021. Dr. Martens pointed to the Kaiser study, along with the landmark MOBILE study, as a driving force for the ADA’s 2022 Standards of Care, which championed CGM use for individuals on basal-only therapy. Dr. Simonson then took over and highlighted a systematic review (Tsapas et al., 2020) showing that while metformin may be a fine first-line therapy for individuals with low CV risk, those at a higher CV risk might also benefit from a front-line prescription for an SGLT-2 or GLP-1.

Real-world diabetes technology: What do we have? Who are we missing?

• The esteemed Dr. Laurel Messer (Barbara Davis Center) presented on the real-world implementation of diabetes technology, but not without first acknowledging the contributions of the woman scientists in the room who have played pioneering roles in the field. Dr. Messer kept her presentation concise and focused, with a qualitative and quantitative data review to power her central thesis that everyone in the room can play a role in narrowing technology disparities across socioeconomic status (SES). To start, Dr. Messer touched on an article also referenced by Dr. Maahs (Holl et al., 2021), finding a slight increase in pump utilization across SES in Germany, and a greater SES disparity in the US compared to Germany. Dr. Messer explained that this data is “actionable and interesting” because after adjusting the A1c data by technology use, it shows that glycemic outcomes have not been getting worse over time. As such, Dr. Messer stressed the importance of expanding access to diabetes technology, especially in high-risk, marginalized populations, given that data unequivocally show that everyone, regardless of SES, can clinically benefit from diabetes technology. On the qualitative side, Dr. Messer pointed to a study with semi-structured interviews among 40 young adult type 1s, 72% of whom are Medicaid beneficiaries (Miller et al., 2021). Notably, this paper highlights how providers can act as gatekeepers, dictating who does and does not receive technology in a way that can be influenced by personal biases. In a more optimistic vein, Dr. Messer described a participant response highlighting the power of provider optimism and how it can inspire people with diabetes to start using technology.

Diabetes technology and the human factor

• Dr. Alon Liberman once again presented the ATTD Yearbook chapter on “Diabetes Technologies and the Human Factor,” reminding attendees that patients still play an “essential role” in the successful use of diabetes technology. This year, Dr. Liberman focused on the concerning cross-Atlantic data that stratified A1c, CGM use, and pump use by socioeconomic status in the DPV and T1D Exchange database, which was published in Diabetes Care. Overall, the study found that disparities were present in both geographies, but that they were far worse in the US. These data were also presented in other chapter presentations, but as Dr. Phillip noted, it is important enough that it should be presented repeatedly. The second study that Dr. Liberman presented, which was published in Pediatric Diabetes, looked at the challenges faced by children using technology in schools in working with school nurses to manage their diabetes and devices, a major issue that will be of increasing importance as more and more children use diabetes technology.

Immune intervention for type 1 diabetes

• Dr. Bimota Nambam (Johns Hopkins) once again took on the advancements in immune intervention for type 1 diabetes chapter of this year’s ATTD Yearbook. This year, Dr. Nambam focused on three studies on three immune interventions for type 1 diabetes. The first is one with which many will be familiar: a trial evaluating teplizumab that was published in Science Translational Medicine and found that teplizumab improved and stabilized beta cell function in antibody-positive, high-risk individuals and doubled the median time to type 1 diagnosis relative to the control group. Compared to the second two studies presented this was a marked success. Second, Dr. discussed five-year follow-up results evaluating a combination immunomodulatory agent treatment, which was published in Diabetes and found no statistical improvement in A1c or beta cell function markers in those with newly diagnosed type 1 diabetes after five years relative to the control group. Finally, Dr. Nambam closed with the results of an RCT evaluating an oral insulin as an agent to prevent autoimmunity in very young children, which was published in Diabetologia. Unfortunately, the study found no significant difference between the treatment and control arm at six months; however, those with the INS genotype saw antibody responses to insulin in the treatment group relative to the control group.

New medications for the treatment of diabetes

• Dr. Viral Shah (Barbara Davis Diabetes Center) outlined recent advancements in the therapeutic treatment of diabetes for both type 1 and type 2. Starting with type 1 diabetes, Dr. Shah discussed SGLT inhibitors noting that while there is an increased risk of DKA for people with type 1 diabetes on SGLT inhibitors, they have also been shown to drive improvements in glycemic management. Dr. Shah also highlighted a phase 2 trial on the use of a glucagon-receptor antibody (volagidemab), which demonstrated a small decrease in A1c by ~0.2-0.3% and a reduction in insulin requirement, but did have some concerning side-effects including increased blood pressure. Turning to type 2 diabetes, Dr. Shah focused solely on tirzepatide emphasizing the results of SURPASS-2, which demonstrated A1c reductions up to 2.5% among patients on the highest dose (15mg). Dr. Shah also discussed emerging medications for the treatment of diabetes-related complications highlighting finerenone and the FIDELIO-DKD trial for diabetic kidney disease and anti-VEGF treatment in diabetic retinopathy.