Trial results for an AI Advisor-driven at-home closed loop system for Type 1 Diabetes in young children were posted on ClinicalTrials.gov on 2025-09-15, showing an increase in time in target glucose range from a mean of 63% to 70%.
Background
Type 1 Diabetes requires continuous management of blood glucose levels, a challenging task, especially in young children. Advanced technologies like closed-loop control (CLC) systems, often referred to as artificial pancreases, have transformed diabetes care by automating insulin delivery based on continuous glucose monitor (CGM) readings. However, optimizing these systems for individual patient needs, particularly in pediatric populations, remains an area of active development. The integration of AI-based advisor systems aims to further refine parameter adjustments, potentially improving glycemic control and reducing the burden on patients and caregivers.
Trial design
This completed study, identified as Phase NA, enrolled 33 participants with Type 1 Diabetes. The trial investigated the safety and exploratory glycemic control data from using an at-home closed-loop control (CLC) system (t:slim X2 with Control-IQ Technology) with periodic parameter adjustments driven by an AI-based Advisor system in young children. Participants used the study system in closed-loop mode for eight weeks. The main endpoints aimed to assess the safety and efficacy of the AI-driven pump parameters, comparing outcomes to a baseline period.
Key results
The trial results compared outcomes during the AI Advisor-driven at-home closed loop system period against a baseline period. Key measurements included both safety and hierarchical efficacy endpoints:
- For the safety endpoint measuring the percentage of time with CGM-measured glucose levels below 54 mg/dl, the mean was 0.3% (Standard Deviation: 0.3%) during the baseline period and remained at 0.3% (Standard Deviation: 0.2%) with the AI Advisor system.
- Another safety endpoint, the percentage of time with CGM-measured glucose levels below 70 mg/dl, showed a mean of 13% (Standard Deviation: 10%) at baseline, which decreased to a mean of 9% (Standard Deviation: 5%) with the AI Advisor system.
- Regarding hierarchical efficacy endpoints, the percentage of time spent in the target glucose range (70-180 mg/dl) increased from a mean of 63% (Standard Deviation: 15%) during baseline to a mean of 70% (Standard Deviation: 8%) with the AI Advisor system.
- The mean CGM-measured glucose level decreased from 168 mg/dl (Standard Deviation: 28 mg/dl) at baseline to 157 mg/dl (Standard Deviation: 14 mg/dl) with the AI Advisor system.
- The percentage of time with CGM-measured glucose levels above 180 mg/dl decreased from a mean of 13% (Standard Deviation: 10%) at baseline to a mean of 9% (Standard Deviation: 5%) with the AI Advisor system.
- The percentage of time with CGM-measured glucose levels below 70 mg/dl, also listed as an efficacy endpoint, decreased from a mean of 2% (Standard Deviation: 1.4%) at baseline to a mean of 1.7% (Standard Deviation: 0.9%) with the AI Advisor system.
What this means
The results suggest that the AI Advisor-driven at-home closed loop system can improve glycemic control in young children with Type 1 Diabetes. The increase in time within the target glucose range to a mean of 70% and the reduction in mean glucose to 157 mg/dl are clinically meaningful improvements. Furthermore, the observed decrease in time spent in hyperglycemia (above 180 mg/dl) and hypoglycemia (below 70 mg/dl) indicates enhanced safety and stability of glucose levels. These findings imply that AI-driven parameter adjustments can optimize closed-loop systems, potentially easing the management burden for families and leading to better health outcomes for pediatric patients.
Source
The information regarding these trial results was obtained from ClinicalTrials.gov, a public database of clinical studies. The results for the study NCT06017089, titled "The Pediatric Artificial Pancreas Automated Initialization Trial", were posted on 2025-09-15 on clinicaltrials.gov.