P-144: Real-world experience of pediatric users of the new sensor with the advanced hybrid closed loop (AHCL) system in the United States
J Shin, M Liu, f niu, c Rodriguez, V Tran, T Cordero, R VigerskyAbstract
Introduction
The Instinct sensor(INS), made by Abbott, was recently integrated into the MiniMed™ 780G (MM780G) system. This provides U.S. pediatric MM780G system users with a new CGM option, necessitating a real-world evaluation of performance and CGM compatibility during transition from the Guardian™ 4 sensor (G4S).
Objectives
Study aimed to evaluate the performance of the MM780G system with the INS in a pediatric population (≤17 years). Key focus areas included the impact of Recommended Optimal Settings (ROS) on glycemic outcomes and the effects of transitioning from G4S to INS.
Methods
CareLink™ data were retrospectively analyzed for 498 pediatric users in the U.S. as of March 2026. Outcomes were also stratified by ROS (100 mg/dL target; 2h active insulin time) versus non-ROS configurations. Also, a subset of 232 users who transitioned from G4S to INS (with ≥ 30 days of data on each sensor) was evaluated for interchangeability.
Results
The pediatric INS cohort maintained a mean Time in Range (TIR) of 65.0%. Stratification revealed that ROS was a primary driver of superior outcomes; ROS users achieved a TIR of 68.7%, which was significantly higher than the 63.8% observed for non-ROS users. Time Below Range (TBR) was similar between subgroups (2.5% vs. 2.7%). In the pre/post-switch analysis, TIR remained stable between G4S and INS (65.9% vs. 67.2%). While TBR was numerically higher with INS (2.5% vs. 1.4%), it remained well within the ADA-recommended <4% threshold. Notably, INS demonstrated good system engagement, with higher sensor wear-time (95.0% vs. 89.6%) and Auto Mode utilization (94.2% vs. 88.4%) compared with the G4S.
Conclusions
Real-world data confirm that the INS and G4S are clinically interchangeable within the MM780G system. The longer wear-time of the INS appears to enhance Auto Mode uptime. However, glycemic optimization is primarily dependent on the MM780G AHCL algorithm and the adoption and use of Recommended Optimal Setting rather than specific sensor.