Automated Insulin Delivery

Could AID Transform Type 2 Diabetes Care?

Posted by

0


While AID has traditionally been used in type 1 diabetes, new data suggests this technology has many of the same benefits in type 2 – namely, improving time in range and A1C while reducing hypoglycemia. Plus, AID dramatically simplifies blood sugar management. 

From continuous glucose monitoring (CGM) to automated insulin delivery (AID) systems, diabetes technologies that began as innovations for people with type 1 diabetes are slowly beginning to reach people with type 2.

For instance, many insurance companies now cover CGMs for people with type 2 diabetes who take insulin as well as those who are not on insulin but have a history of hypoglycemia. And earlier this week, the FDA approved Stelo by Dexcom, a CGM designed specifically for people with type 2 diabetes who are not taking insulin.

However, less progress has been made with reimbursement by insurance companies for AID. Off-label use of AID drew significant attention at the ATTD 2024 conference, with presenters highlighting the benefits for many people with diabetes across a range of settings and systems.

AID improves time in range across different systems and settings

Research shows that AID leads to many of the same benefits in type 2 diabetes as in type 1 diabetes: improved time in range, reduced hypoglycemia, and reduced A1C. Importantly, these benefits were consistent across different study settings and regardless of which AID system was used.

A study of 30 Tandem Control-IQ users with type 2 diabetes found that time in range increased by about 15% from 56% at baseline to 71% at six weeks. This translates to an increase of 3.6 hours per day spent in range.

Dr. Anders Carlson, diabetes medical director at the International Diabetes Center in Minnesota, said this finding is in line with studies in type 1 diabetes as well as the time in target range guidelines for type 1 diabetes.

Outside of clinical trials, research suggests that the benefits of AID extend to people with type 2 diabetes in the “real world.”

In a study presented at ATTD, MiniMed 780G users were able to achieve 71-75% time in range outside of a clinical trial, again meeting the targets for diabetes. “This is really compelling evidence that in a real-world setting, this AID system can work for people with type 2 diabetes,” Forlenza said.

Participants who used the recommended MiniMed 780G settings (i.e. the lowest glucose target) achieved a time in range of 80%.

For Carlson, this finding raises an important question – what are the optimal settings for AID in type 2 diabetes? For instance, since low blood sugar (hypoglycemia) is less of a concern, it may be beneficial to have more aggressive targets from the get-go.

Another study investigated the Omnipod 5 AID system in 24 participants with type 2 diabetes, finding strong improvements in time in range with minimal hypoglycemia. Among those on MDI, time in range increased from 43% at baseline to 58% at six months. Participants on basal insulin only saw even larger improvements in time in range, from 31% at baseline to 65% at six months.

Dr. Anne Peters, professor of medicine at USC, also highlighted reductions in total daily insulin dose among participants on MDI – yet another way in which AID could simplify type 2 diabetes management.

How might combining AID with GLP-1s and SGLT-2s affect glucose levels?

Growing use of GLP-1 receptor agonists, SGLT-2 inhibitors, and diabetes technology poses new questions for the future of diabetes care. That is, how might the combination of technology and medications optimize outcomes for people with type 2 diabetes?

In the Omnipod 5 study, half of the patients were also taking a GLP-1 or SGLT-2. Overall, Omnipod users taking a GLP-1 or SGLT-2 saw greater improvements in time in range compared to those who were only taking insulin. Participants in the GLP-1 or SGLT-2 group saw a 24% increase in time in range from 28% at the start of the study to 62% at eight weeks. Meanwhile, participants not using a GLP-1 or SGLT-2 improved their time in range by 18%, from 35% at baseline to 53% at eight weeks.

Carlson said this finding suggests that combining GLP-1s or SGLT-2s with AID could potentially lead to even better glycemic control than AID alone – though formal studies will be needed to test this hypothesis.

Similarly, Dr. Gregory Forlenza, associate professor of pediatric endocrinology at the University of Colorado, noted the ability of GLP-1s to reduce insulin needs. Combining these powerful medications with AID may help people with type 2 diabetes improve glycemic control and lose weight. It’s possible these improvements could even help people work toward diabetes remission.

What about AID for older adults with type 2 diabetes?

Starting insulin can be challenging for people of all ages, but it can be especially complex for older adults or disabled people with type 2 diabetes who receive home care.

Elderly people have a higher risk of severe hypoglycemia and hypoglycemia or ketoacidosis. Diabetes management for older adults can also be complicated by impaired cognition or dementia, reduced mobility, and difficulty accessing care.

In this context, the CLOSE AP+ study investigated AID assisted by nurses in people with type 2 diabetes unable to manage their own multiple daily injections (MDI) at home. CLOSE AP+ tested Control-IQ technology in 25 participants who had an average age of 70 years.

At 12 weeks, time in range improved significantly, from 37% to 63%. Time below range was less than 1%, while time above range was under 10%. Overall, Reznik highlighted that a majority of participants reached the American Diabetes Association guidelines for older people with diabetes. These guidelines recommend:

  • At least 50% time in range (70-180 mg/dL)
  • Less than 1% time below range (<70 mg/dL)
  • Less than 10% time above range (>250 mg/dL)

It’s also worth noting that participants using Control-IQ technology saw a significant 1.3% reduction in A1C. Over 90% of participants reached an A1C of less than 8% by the end of the trial, without any increase in severe hypoglycemia. Dr. Yves Renzik, professor of endocrinology at CHU Caen Normandy in France, also highlighted high patient confidence and high nurse satisfaction with the AID system in this study.

Ultimately, the CLOSE AP+ study showed that AID can be used safely in people with type 2 diabetes who require home nursing care. This confirms the benefits of AID extend beyond the “standard” person with type 2 diabetes to older adults and people with disabilities.

The bottom line

Numerous presentations at ATTD 2024 demonstrated that AID is safe and effective for people with type 2 diabetes. Both clinical trials and real-world data show that this technology increases time in range and improves A1C while minimizing hypoglycemia.

“I want to emphasize that across a wide variety of real world and clinical trial evidence sets, and across very different AID systems, everyone is either doing a great job hitting a goal for time in range or achieving a massive improvement in glucose control,” Forlenza said. He noted that AID leads to time in range increases of 15% to 24% in people with type 2 diabetes, nearly double the improvements typically seen in type 1 diabetes.

However, several questions remain to be answered regarding optimal settings, bolusing, and the potential of AID when combined with GLP-1s and SGLT-2s. Carlson highlighted the following areas for further research:

  • Are people on MDI the only candidates for AID? Or could AID be used in all people with type 2, regardless of their insulin needs and whether or not they’re meeting glycemic goals?
  • Is previous experience with technology necessary for successful use of AID in people with type 2 diabetes?
  • Does AID help with diabetes self-management (such as carb awareness)?
  • What role will primary care providers provide in supporting AID in this population?

Beyond glycemic data, it’s also important to consider user experience with AID. Overall, the data suggests that people with type 2 diabetes had good satisfaction and confidence in using these systems. Even people who hadn’t previously used diabetes devices reported a positive experience with AID, Peters noted.

“I honestly wasn’t sure my patients would like AID – many were technology-naive people,” Peters said. “But they loved it and they wanted to stay on it because they felt it improved their glycemic control.”

The Future of Diabetes Care – Artificial Intelligence, Telemedicine, and Automated Insulin Delivery

Posted by

0


A fascinating session at the EASD 2022 conference on emerging technologies sheds light on where we are with AID and telemedicine, and what leading researchers in diabetes believe is coming next in diabetes management.

Healthcare is rapidly evolving, and now more than ever, robots and artificial intelligence have gone from science fiction to critical components of diabetes management. At the EASD 2022 conference in Stockholm, Sweden, researchers further explored this concept in a session titled, “A New Hope or Strange New Worlds: Submerging diabetes into emerging technologies.”

Clinical Decision Support

Dr. Moshe Phillip, head of the Institute of Endocrinology and Diabetes at Schneider Children’s Medical Center of Israel, began by demonstrating how continuous glucose monitors (CGMs) represent a paradigm shift in diabetes technology. “CGM is the most important tool in the last 20 years,” he said. 

Philip added that automated insulin delivery technology has resulted in remarkable increases in time in range (TIR) and lower A1C levels.

Phillip explained how these technologies intersect with artificial intelligence to create a Clinical Decision Support System (CDSS). He pointed out that diabetes treatment guidelines are “so complex that it is very difficult to follow.” 

The CDSS incorporates data from technologies such as CGMs, insulin pumps, smart connected pens, and even self-monitored blood glucose meters. This data is analyzed by artificial intelligence algorithms that detect glucose patterns and insulin dosing events, ultimately providing an in-depth visualization of an individual’s health data. 

“Within a split second, physicians get information on how to tailor the therapy for each individual patient,” Philip said. Like knocking on your colleague’s door and asking for a second opinion, the CDSS is intended to be used as a supportive tool. 

With a more convenient visual representation of health data, physicians do not need to spend as much time sifting through pages of data and can spend more time facilitating a discussion with their patients.

Advantages and Challenges of Telemedicine

Although telemedicine has been on the rise for several years, the COVID pandemic led to a rapid acceleration of its adoption. Dr. Richard Holt, professor in diabetes and endocrinology at the University of Southampton, said, “Overnight, we were forced to adapt to telemedicine because face-to-face was no longer safe with the pandemic.”

People with diabetes appreciated the increased convenience and flexibility of telemedicine, while healthcare providers benefited from fewer missed appointments and cancellations. Additionally, not having to wait in the waiting room increased people’s privacy and reduced the risk of the spread of infections in the clinic. 

However, Holt pointed out that the shift to telemedicine came with various challenges. He cited a study that suggested around 60,000 cases of diabetes might have gone undiagnosed in the United Kingdom because of the COVID pandemic. There were also significantly fewer A1C tests and foot screenings during the pandemic. “I don’t know how to examine the foot over the telephone or internet,” said Holt. 

Some individuals preferred in-person visits as it made it easier to engage with their healthcare team. Several people found it more stressful to use telemedicine, demonstrating that we cannot assume the technology will be convenient and comfortable for everyone. Holt noted that telemedicine is less suitable for those who are very young or very old, those with complex health needs, and those that require a physical examination. Holt concluded by explaining that while telemedicine can be used to deliver effective diabetes care, “it cannot replace all in-person consultations.”

Dr. Steven Russell, associate professor of medicine at Harvard Medical School, reviewed the state of automated insulin delivery and hinted at potential smart insulin options in the future. At the start of his talk, Russell acknowledged that there is a severe shortage of endocrinologists in the United States, and most primary care physicians feel uncomfortable managing people with type 1 diabetes who are using an insulin pump.

Hybrid closed-loop systems partially automate insulin dosing, requiring only manual mealtime boluses and occasional correction boluses. He went into detail about the Beta Bionics iLet Bionic Pancreas and how it improved time in range and lowered A1C levels. He also cited a study that showed how an AID system featuring automatic microdosing of glucagon may prevent or reduce severe hypoglycemia. Russell called for larger and longer studies on these bi-hormonal (insulin and glucagon) systems.

He concluded his talk by briefly discussing smart insulins – or an insulin that is activated when blood glucose rises and deactivates if glucose levels go below 50. This smart insulin would be “already where it is needed, it just needs to be activated or deactivated.” 

Russell noted that “although we are not quite there yet with smart insulin, we are getting closer.”

FDA approves first smartphone app for bolus insulin delivery

Posted by

0


The FDA has approved the first smartphone application capable of initiating insulin delivery for insulin pump wearers, according to a company press release.

Tandem Diabetes Care has received clearance to offer a bolus insulin dosing feature as part of its t:connect mobile application. The app, which connects to the t:slim X2 insulin pump, will include a new feature giving users the ability to program and cancel bolus insulin requests through their smartphone.

Sign outside FDA HQ in Washington, DC.

“This FDA clearance further validates our commitment to innovation and the diabetes community by providing one of the most requested feature enhancements,” John Sheridan, president and CEO for Tandem Diabetes Care, said in the release. “With the improvements in diabetes management provided by Tandem’s Control-IQ technology, giving a meal bolus is now the most common reason a person interacts with their pump, and the ability to do so using a smartphone app offers a convenient and discrete solution.”

The t:connect mobile app, which is available for iOS and Android devices, provides 24-hour glucose trends, pump status changes and insulin therapy data for users who pair it with the t:slim X2 insulin pump. The insulin pump can function independently from the app, giving users the ability to view pump therapy data, program requests and cancel bolus insulin requests from the pump.

The bolus insulin dosing feature will be offered at no additional cost for new t:slim X2 insulin pump customers as well as existing in-warranty customers through a remote software update, according to the company. The update will be rolled out in a series of limited launch groups in the spring. According to the press release, limited launch participants have already been selected. An expanded launch of the new feature is expected later in the summer.

No changes in glucose, insulin delivery during menstrual cycle with artificial pancreas

Posted by

0


A small group of women with type 1 diabetes using a hybrid closed-loop insulin delivery system maintained similar glycemic levels and insulin delivery throughout their entire menstrual cycle, according to study data

In findings published in Diabetes Technology and Therapeutics, a group of 16 women with type 1 diabetes had no significant changes in glycemic control during the luteal phase, menstrual phase and the rest of their menstrual cycle. The findings are the first step toward exploring the potential benefits of using a hybrid closed-loop system to optimize control in women struggling with glycemic control during certain phases of the menstrual cycle, according to Carol Levy, MD, clinical director of the Mount Sinai Diabetes Center.

Carol Levy, MD
Levy is clinical director of the Mount Sinai Diabetes Center.

“What’s been a challenge for clinicians is how to help manage patients who struggle with glycemic changes around the menstrual cycle and provide them guidance,” Levy told Healio. “This is a subanalysis of a larger study. We thought it would be interesting to see what insulin delivery and glycemic metrics we would get around the menstrual cycle. We took a small group of women within the larger study who agreed to do it. What we found was there weren’t marked changes with a hybrid closed-loop system.”

Levy and colleagues enrolled 16 menstruating women with type 1 diabetes from the International Diabetes Closed Loop trial to participate in a substudy (69% white, mean age, 31 years). Participants installed a tracking application on their personal phone to track menstrual cycle phases. All participants used a t:slim X2 insulin pump with Control-IQ technology (Tandem Diabetes Care) and the Dexcom G6 continuous glucose monitoring system. CGM metrics and insulin delivery during the mid-late luteal phase, menstrual phase and the rest of the menstrual cycle were analyzed for the cohort.

Data was reported for a mean of 145 days and six menstrual cycles. The cohort had a mean 24-hour glucose level of 161 mg/dL during the menstrual phase, 165 mg/dL during the luteal phase and 159 mg/dL during the rest of the cycle. Mean time in range was 69% in the menstrual phase, 67% during the luteal phase and 69% during the rest of the cycle.

Insulin delivery was unchanged during the menstrual cycle, with median basal rates remaining at 0.28 U/kg during all three phases. Median daily bolus insulin delivery rates were 0.31 U/kg during the luteal phase and 0.29 U/kg during the menstrual phase and rest of the cycle.

“What’s unique about this population is they were all well controlled because they were on the closed-loop system, so perhaps improved glycemic control may be able to blunt some of the disturbances [in insulin sensitivity] some women see during their menstrual cycle,” Levy said. “Women are also sometimes chasing these high blood sugars that they see around the menstrual cycle. The automated system may provide some benefit and not require much larger differences in insulin delivery, but we do not yet know all of the answers.”

Levy said future studies should focus on women who report having challenges maintaining a stable glycemic control throughout their menstrual cycle and analyze the effects of a hybrid closed-loop system.

“It may not even be that putting them on a closed-loop system will resolve their fluctuating glucose levels,” Levy said. “The closed-loop system with adjustments and other settings may really help them.”

Automated Insulin Delivery (Artificial Pancreas, Closed Loop)

Posted by

0


artifiical pancreas

 

The development of automated insulin delivery has many names – artificial pancreas, hybrid closed loop, Bionic Pancreas, predictive low glucose suspend – but all share the same goal: using continuous glucose monitors (CGMs) and smart algorithms that decide how much insulin to deliver via pump. The goal of these products is to reduce/eliminate hypoglycemia, improve time-in-range, and reduce hyperglycemia – especially overnight.

See below for an overview of the automated insulin delivery field, focused on companies working to get products approved. Do-it-yourself automated insulin delivery systems like OpenAPS and Loop are not included here, though they are currently available and used by a growing number of motivated, curious users.

We’ve also included helpful links to articles on specific product and research updates, as well as some key questions.

Who is Closing the Loop and How Fast Are They Moving?

Below we include a list of organizations working to bring automated insulin delivery products to market – this includes their most recently announced public plans for pivotal studies, FDA submissions, and commercial launch. The organizations are ordered from shortest to longest time to a pivotal study, though these are subject to change. This list excludes those without a commercial path to market (e.g., academic groups). The first table focuses on the US, with European-only systems listed in the second table.

Updated: November 4, 2017

US Products

Company / Organization Product Latest Timing in the US
Medtronic MiniMed 670G/Guardian Sensor 3 – hybrid closed loop that automates basal insulin delivery (still requires meal boluses) FDA-approved and currently launching this fall to ~35,000 Priority Access Program participants in the US. Pump shipments to non-Priority Access customers will start in October, with sensors and transmitters to ship by the end of 2017 or early 2018. Medtronic is experiencing a global CGM sensor shortage that won’t resolve until spring 2018.
Tandem t:slim X2 pump with built-in predictive low glucose suspend (PLGS) algorithm; Dexcom G5 CGM

t:slim X2 pump with built-in Hypoglycemia-Hyperglycemia Minimizer algorithm; Dexcom G6 CGM (including automatic correction boluses)

 Launch expected in summer 2018. Pivotal trial now underway, with FDA submission expected in early 2018.

Launch expected in the first half of 2019. Pivotal trial to begin in the first half of 2018.
Insulet OmniPod Horizon: pod with built-in Bluetooth and embedded hybrid closed loop algorithm, Dash touchscreen handheld, and Dexcom G6 CGM

User will remain in closed loop even when Dash handheld is out of range

 Launch by end of 2019 or early 2020, with a pivotal study in 2018
Bigfoot Biomedical Smartphone app, insulin pump (acquired from Asante), and a next-gen version of Abbott’s FreeStyle Libre CGM sensor (continuous communication)

The smartphone is expected to serve as the window to the system and complete user interface

 Launch possible in 2020, with a pivotal trial expected in 2018
Beta Bionics Bionic Pancreas iLet device: dual chambered pump with built-in algorithm; hybrid or fully closed loop; insulin-only or insulin+glucagon; custom infusion set, Dexcom CGM

Likely to launch as insulin-only product, with glucagon to be optionally added later

Currently using Zealand’s pumpable glucagon analog

 Insulin-only: possible US launch in the first half of 2020, with a pivotal trial to start in the beginning of 2019.

Insulin+glucagon (bihormonal) pivotal trial expected to start in the beginning of 2019. Timing of FDA submission and launch depend on a stable glucagon, among other things.

European Products

Company / Organization Product Latest Timing in Europe
Medtronic MiniMed 640G/Enlite Enhanced – predictive low glucose management

MiniMed 670G/Guardian Sensor 3 – hybrid closed loop that automates basal insulin delivery (still requires meal boluses)

 Currently available in Europe

No timing recently shared. Approval was previously expected in summer 2017
Diabeloop Diabeloop algorithm running on a wireless handheld, Cellnovo patch pump, Dexcom CGM Pivotal trial expected to complete in February/March 2018. Possible European launch in 2018
Roche, Sensonics, TypeZero Will use Senseonics’ 180-day CGM sensor, Roche pump and TypeZero algorithm Pivotal trial expected to begin in Europe in early 2018
Cellnovo, TypeZero Cellnovo patch pump with integrated TypeZero algorithm; presumably a Dexcom CGM Aims for a 2018 European launch. No pivotal trial details shared

Helpful Links

Medtronic: MiniMed 670G

Tandem

Insulet

Bigfoot

Beta Bionics

Test Drives:

test drive – UVA’s Overnight Closed-Loop Makes for Great Dreams. Kelly participates in UVA’s overnight closed loop trial and reports back on an incredible opportunity for the field to move fast, reduce anxiety, and beat timelines.

test drive – Kelly and Adam take UVA’s DiAs artificial pancreas system home 24/7 for a three-month study. Their key takeaways, surprises, and next steps.

Key Questions for the Artificial Pancreas

Are patient expectations too high? If we expect too much out of first-generation artificial pancreas systems – e.g., “I don’t have to do anything to get a 6.5% A1c with no hypoglycemia” – we might be disappointed. Like any new product, early versions of the artificial pancreas are going to have their glitches and shortcomings. Undoubtedly, things will improve markedly over time as algorithms advance, devices get more accurate and smaller, insulin gets faster, infusion sets improve, and we all get more experience with automated insulin delivery. But it takes patience and persistence to weather the early generations to get to the truly breakthrough products. We would not have today’s small insulin pumps without the first backpack-sized insulin pump; we would not have today’s CGM without the Dexcom STS, Medtronic Gold, and GlucoWatch; we would not be walking around with smartphones were it not for the first brick-sized cellphones. Our research trial experience with automated insulin delivery recalibrated our expectations a bit – these systems are going to be an absolutely terrific advance for many patients, but they will not replace everything out of the gate. Let’s all remember that devices need to walk first, then run, and it’s okay if the first systems are more conservative from a safety perspective.

What fraction of patients will be willing to wear some type of automated insulin delivery system? Right now, many estimate that ~30% of US type 1’s wear a pump, and about 15% to 20% wear CGM. There are a lot of reasons why that may be the case, including cost, hassle, no perceived benefit, no desire to switch from current therapy, wearing a device on the body, alarm fatigue, etc. Will automated insulin delivery address enough of these challenges to expand the market?

Will healthcare providers embrace automated insulin delivery? Today, healthcare providers lose money when they prescribe pumps and CGM – they are very time consuming to train, prescribe, and obtain reimbursement for. We need to make sure that automated insulin delivery systems make providers’ lives easier, not more complicated.

Will there be a thriving commercial environment and reimbursement? It’s extremely expensive to develop and test closed-loop systems, and companies will only develop them if there is a commercial environment that supports a reasonable business. Reimbursement is a major part of that, and it’s hard to know if insurance companies will pay for closed-loop systems for a wide population of patients. We are optimistic that reimbursement will be there, especially if systems can simultaneously lower A1c, reduce hypoglycemia, and improve time-in-range.

What’s the right balance between automation and human manual input? The holy grail is a fully-automated, reactive closed loop that requires no meal or exercise input. But insulin needs to get faster to make that a reality. For now, daytime systems need to deal with balancing human input with automation, and there’s an associated patient learning curve. How much should automated insulin delivery systems ask patients to do? How do we ensure patients do not forget how to manage their diabetes (“de-skilling”) as systems grow in their automation abilities?

Insulin-only or insulin+glucagon? Ultimately, we believe that the question is partially one of patient preferences. There will be some patients who may want the extra glycemic control offered by the dual-hormone approach and will be willing to accept a bit more risk or a more aggressive algorithm. An insulin+glucagon system could be helpful for those with hypoglycemia unawareness, and if such a system makes it to the market, some patients will certainly want to give it a try. We believe a range of options is a good thing for people with diabetes, since all systems and products have pros and cons. Ultimately, cost considerations may present the largest factor in adoption. An insulin+glucagon system certainly brings multiple cost elements to consider – a second hormone, a dual-chambered pump, custom infusion sets, potentially higher training, etc. It’s hard to know at this point how the relative costs/benefits will exactly compare to insulin-only systems.