'Mix and Match' Options Emerging in Automated Insulin Delivery

Miriam E. Tucker

June 09, 2019

SAN FRANCISCO — As automated insulin delivery is rapidly evolving, interchangeability between insulin pumps and sensors on the various platforms is emerging as a key component to moving the field forward.

Here at the American Diabetes Association (ADA) 2019 Scientific Sessions, many attendees who follow the technology were surprised by the announcement that Medtronic will be working with the nonprofit Tidepool to create an interoperable automated insulin pump system, a category the US Food and Drug Administration (FDA) now calls "alternate controller enabled (ACE)" pumps.

Currently, Medtronic manufactures the only commercially available hybrid closed-loop system (previously known as the artificial pancreas) and is the only company to manufacture both the insulin pump and continuous glucose monitor (CGM) components of such automated insulin delivery systems.

Tidepool is a data hub for patients and clinicians to combine and view data from insulin pumps, CGMs, and blood glucose meters. It also stores information about meals, exercise, and other daily events. As a next step, the organization is now developing the Tidepool Loop app, which would serve as a platform allowing interoperability between various diabetes devices. 

"Pairing our future Bluetooth-enabled MiniMed ACE pump with Tidepool Loop would enable an FDA-approved interoperable system — with pump and CGM components — that may be mixed and matched with the Tidepool Loop app as the person with diabetes chooses," Medtronic announced on June 6, the day before the conference.

It's Tough Being First...

Here at the meeting, an oral abstract session on automated insulin delivery held on June 8 included a presentation from Stanford University on real-world clinical experience with Medtronic's hybrid MiniMed 670G system in which — similar to a report at ENDO 2019: The Endocrine Society Annual Meeting in March — clinical experience with this first commercially available artificial pancreas has been less than ideal, with many patients abandoning the system and frequently citing difficulties with the sensor as the reason.

"It's always tough being first, from the standpoint of the 670G, and this cooperation [with Tidepool] is going to give folks all sorts of other options," Rayhan Lal, MD, of Stanford University, California, told Medscape Medical News. "Other companies are starting to collaborate, and I think to stay in the space you have to cooperate with others in the field," he added.

Indeed, session moderator Alanna Weisman, MD, University of Toronto, Ontario, Canada, told Medscape Medical News, "Interchangeability is a big issue...I think patients would like to be able to mix and match and choose the devices that work best for them."

Weisman said she was surprised by the Medtronic announcement, but that "it's very exciting."

Also presented during the same abstract session were new data on the third generation of Beta Bionics' investigational iLet system, including its use with two different sensors — the Dexcom G5 and implantable Eversense (Senseonics).

Real-World Results Differ From Research

Lal presented data from a 1-year prospective observational study of 79 adult and pediatric patients (aged 9-61 years) who started using the 670G system at Stanford between May 2017 and May 2018. Most (72%) had previously used a Medtronic pump, 51% had previously used a Dexcom sensor, and 33% had previously used a Medtronic CGM.    

The proportions discontinuing the "auto mode" function of the 670G rose over time, from just 1% at week 1 to 40% at 6 months to 46% at 1 year. Compared to those who continued using auto mode, those who discontinued were significantly younger (22.3 vs 31.5 years; P = .02) and had been using the Medtronic Guardian 3 sensor for less time prior to entering auto mode (P = .001).

Of note, although it wasn't significant due to low numbers, participants who were using a Dexcom along with the 670G – and feeding the numbers from the Dexcom into the 670G system – were also more likely to discontinue auto mode (31% vs 13%; 8 vs 4 patients; P = 0.11).

"Sensor issues" accounted for 60% of the reasons listed for abandoning auto mode at 1 year, including need for multiple daily calibrations, sensor alerts, and systems automatically exiting auto mode and requiring additional fingersticks. Other cited reasons, such as problems obtaining supplies and fear of hypoglycemia, were less common (17% and 10%, respectively). 

"Education and adequate preparation are crucial in setting realistic expectations for closed-loop systems. A focus on usability and human factors is necessary to ensure patients stay on treatment," Lal concluded.

In separate press releases issued on June 8, Medtronic announced enrolment of the first study participants in a pivotal trial of its Bluetooth-enabled 780G advanced hybrid closed-loop system, designed to automate the delivery of correction boluses to address current or anticipated high blood glucose levels based on sensor readings, and a next-generation Guardian CGM, designed to improve accuracy and system performance and reduce the number of calibrations.  

iLet Works With Different Sensors, Ultimately to Include Glucagon

Rabab Z. Jafri, MD, a pediatric endocrinologist at Massachusetts General Hospital, Boston, presented the iLet data, the first on safety and efficacy of the Gen3 iLet, "a purpose-built bionic pancreas platform" designed to administer both insulin and glucagon, although the current data are for use with insulin alone.

The iLet uses only body weight to initialize and uses autonomous machine learning to adapt to the individual user. Unlike other closed-loop systems, it doesn't require the user to enter carbohydrate counts, basal rates, or correction factors. It responds to input from the Dexcom G5 or Eversense CGM.

In the current study, 34 adult outpatients were enrolled who had type 1 diabetes. A random-order cross-over was used to compare the insulin-only mode of the iLet to usual care for 7 days each. Mean CGM glucose values were 155 mg/dL while wearing the insulin-only iLet versus 162 mg/dL with usual care (P = .09). Time spent with blood glucose < 54 mg/dL didn't differ significantly (P = .64), but iLet did improve time spent in the 70-180 mg/dL range compared with usual care (70% vs 62%; P = .01). 

Results didn't differ between the 17 patients using the Dexcom G5 and the 17 patients using the Eversense.

Findings from this study have led to improvements in the design of the Gen4 iLet, Jafri noted. 

Separately on June 6, Beta Bionics and Zealand Pharma announced results from the first home-use study of the Gen3 bihormonal configuration of iLet, using Zealand's stable aqueous glucagon analog dasiglucagon.

Ten adults with type 1 diabetes wore the bihormonal and insulin-only iLet configurations for 1 week each. During the bihormonal period, they achieved a mean CGM glucose level of 139 mg/dL on days 2-7 of use compared with 149 mg/dL during the insulin-only period (P < .01).

During the bihormonal period, participants spent 79% of the time with CGM glucose levels in the range of 70-180 mg/dL on days 2-7 of use compared with 71% during the insulin-only period (P < .01).

iLet Available Soon; Patients Want Choices

Principal investigator Steven Russell, MD, Massachusetts General Hospital, Boston, told Medscape Medical News that Beta Bionics plans to start a pivotal study with the Gen4 version — the one they hope to bring to market — next year.

The company anticipates that the insulin-only configuration could be commercially available within 2 years, while the bihormonal version will take longer since it will involve a new drug approval.    

Weisman, who published a meta-analysis of closed-loop systems in 2017, told Medscape Medical News that although her data show the dual-hormone version of iLet has some benefits over the single-hormone version, they also add complexity, so "it's a trade-off."

That's part of the decision-making clinicians will need to help patients within the very near future as these systems reach the market, she noted.

Regarding interchangeability of devices in these systems, Weisman observed: "Patients want choices. I think being able to combine systems to fit their needs makes sense." 

Jafri and Weisman have reported no relevant financial relationships. Lal has reported being a consultant for Abbott Diabetes Care and receives research funding from Medtronic. Russell has reported being on advisory panels for Companion Medical and Unomedical, is a consultant for Flexion Therapeutics, and receives research support from Beta Bionics, MITRE Corporation, Novo Nordisk, and Zealand Pharma. He also has other relationships with ADOCIA, Ascensia Diabetes Care, Lilly Diabetes, Roche Diabetes Care, and Senseonics.

ADA 2019 Scientific Sessions. Presented June 8, 2019. Abstract 80-OR

For more diabetes and endocrinology news, follow us on Twitter and on Facebook.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....