The US FDA has cleared the first updates to the algorithm of the Omnipod 5 automated insulin delivery (AID) system (Insulet Corporation) since its launch in 2022.

Omnipod 5, the only tubing-free pump available in the US thus far, is cleared for people aged 2 or older with type 1 diabetes (T1D) and adults aged 18 and older with type 2 diabetes (T2D). The revisions will allow users to customize target glucose levels down to 100 mg/dL from the current 110 mg/dL and minimize interruptions to automated mode even during prolonged high glucose excursions.

These two features were the most commonly requested enhancements from users, according to an company statement.

Asked to comment, Nicholas B. Argento, MD, diabetes technology director at Maryland Endocrine and Diabetes, Columbia, Maryland, told Medscape Medical News, “I think these are important and useful updates. The target lowered to 100 mg/dL is better because it will allow more time in range without more time below range. Lower is better if it can be done safely.”

In addition, Argento said, “the alterations to reduce time off hybrid closed loop are meaningful. It has up until now thrown people out of loop if they were higher than optimal and not correcting. This was too restrictive. Compared to [the Tandem Control IQ+ systems], which almost never threw people out of loop, it made them look overly conservative, and that frustrated patients. So both of these enhancements are welcome.”

This latest clearance comes as AID system competition is ramping up, with newer products and updates aimed at improving the user experience for people with both T1D and T2D. 

Argento said he’d like to see all AID systems offer some of the newer enhancements, such as the even lower blood glucose target of 87 mg/dL offered by the Twiist (Sequel), approved in March 2024. “That is ideal for nighttime so one can wake up with a lower blood glucose level.”

Another, he said, is a feature expected soon in at least one commercial AID system that would suspend corrections when in exercise or activity mode. “The problem with those corrections is they don’t allow people to carb up before or during exercise without the system giving insulin,” increasing the chance of a hypoglycemic crash, “which is frustrating and forces unwanted calorie intake. This limitation is common to all the current systems as far as I know.”

Regarding the current AID field overall, he said, these are “incremental improvements, but I think we may be on the verge of significant improvements in pattern recognition with algorithms.”

At least in the near future, Argento doesn’t foresee a system that will produce as good control with a fully automated “closed-loop” system as can be achieved now with the user making meal announcement for bolus doses along with automated basal dosing (“hybrid closed-loop”). However, “AI with pattern recognition may get pretty close. It’s an exciting time.”

Argento declared serving on advisory boards for Convatec and Lilly and the speakers bureaus for Dexcom, Tandem, Lilly, Novo Nordisk, and Boehringer Ingelheim. 

Miriam E. Tucker is a freelance journalist based in the Washington DC area. She is a regular contributor to Medscape Medical News, with other work appearing in the Washington Post, NPR’s Shots blog, and Diatribe. She is on X @MiriamETucker and BlueSky @miriametucker.bsky.social.