Diabetes Dialogue: ATTD 2026 Recap

Welcome back to Diabetes Dialogue: Technology, Therapeutics, & Real-World Perspectives!

Advances in artificial intelligence, automated insulin delivery, and sensor integration were central to discussions at the 2026 International Conference on Advanced Technologies & Treatments for Diabetes (ATTD), where clinicians highlighted how emerging technologies may further reduce diabetes management burden while improving glycemic outcomes.

In this episode, hosts Diana Isaacs, PharmD, and Natalie Bellini, DNP, reviewed key highlights from the meeting, with particular emphasis on AI-driven tools and evolving closed-loop systems.

A major theme was the growing role of artificial intelligence, including the concept of “digital twins”—virtual models built from patient-specific data such as glucose trends, insulin dosing, and behavioral inputs. These models allow clinicians and patients to simulate therapy adjustments, such as changes in insulin sensitivity or basal rates, before applying them in practice. Data presented at the meeting suggested use of digital twin modeling improved time in range, whereas providing data feedback alone did not meaningfully change outcomes.

The discussion also examined real-world performance data from newer automated insulin delivery platforms. Early data from the Twiist system demonstrated time in range approaching 76% to 77% across initial user cohorts, with higher time in range observed among individuals using lower glucose targets, albeit with an expected increase in hypoglycemia. Expanded analyses in larger populations showed similarly strong outcomes, reinforcing the importance of target selection and individualized system settings.

Sensor performance and reliability also emerged as a key topic. Investigators presented data suggesting unrecognized infusion set occlusions may contribute to unexplained hyperglycemia, highlighting potential advantages of newer pump technologies designed to detect occlusions more rapidly. In parallel, comparative analyses of sensor integration in hybrid closed-loop systems demonstrated consistent glycemic outcomes across multiple sensor platforms, suggesting algorithm performance, rather than sensor variability alone, plays a dominant role in achieving time in range.

Lastly, hosts shared early data from fully closed-loop systems in type 2 diabetes, including a small trial evaluating automated insulin delivery without meal bolusing. Participants achieved substantial improvements in time in range, increasing from approximately 44% at baseline to 68%, with minimal hypoglycemia. Although not yet achieving traditional glycemic targets, these findings underscore the potential for reducing patient burden while maintaining clinically meaningful glucose control.

Editor’s Note: Isaacs reports disclosures with Dexcom, Abbott, Lilly, Novo Nordisk, Medtronic, Insulet, and others. Bellini reports disclosures with Abbott Diabetes Care, MannKind, Povention Bio, and others.

References:

1. Senseonics. First Ever Real-World Evidence of Eversense 365 Presented at ATTD Demonstrates Sustained Performance and Positive Impact Throughout One-Year of Wear. Senseonics.com. Published March 14, 2026. Accessed March 18, 2026. https://www.senseonics.com/investor-relations/news-releases/2026/03-14-2026-141507610

2. Insulet Corporation. Insulet Presents Promising Study Results for Fully Closed-Loop Automated Insulin Delivery System for Adults with Type 2 Diabetes. Insulet.com. Published March 10, 2026. Accessed March 18, 2026. https://investors.insulet.com/news/news-details/2026/Insulet-Presents-Promising-Study-Results-for-Fully-Closed-Loop-Automated-Insulin-Delivery-System-for-Adults-with-Type-2-Diabetes/default.aspx