(DCP069) ALLEVIATION OF CARBOHYDRATE COUNTING IN TYPE 1 DIABETES WITH AUTOMATED FASTER ASPART AND PRAMLINTIDE CLOSED-LOOP DELIVERY (ARTIFICIAL PANCREAS): A RANDOMIZED CONTROLLED TRIAL

Background: Carbohydrate counting is often used to determine post-prandial insulin needs, but it can be limited by inaccurate calculations, and can negatively impact quality of life and dietary choices. We developed an insulin-and-pramlintide closed-loop system to reduce the need for carbohydrate counting in people with type 1 diabetes.

METHODS AND RESULTS: We conducted a randomized, double-blind, crossover trial comparing 12-day use of (i) insulin-and-placebo closed-loop with carbohydrate counting, (ii) insulin-and-pramlintide closed-loop with simple meal announcement, and (iii) insulin-and-placebo closed-loop with simple meal announcement in a free-living setting. Adults (≥ 18 years) and adolescents (12-17 years) with type 1 diabetes were recruited at the McGill University Health Centre and allocated to an intervention sequence according to a computer-generated randomization code of block size 6. Participants used a closed-loop system integrated into a mobile application which co-administered faster insulin aspart and pramlintide at a 10g:1U ratio, mimicking a co-formulation, and recommended fixed meal boluses upon meal announcements. The primary outcomes of time in target range (3.9-10.0 mmol/L) with a non-inferiority margin of 6.25% and Emotional Burden score of the Diabetes Distress Scale were analysed in participants who completed all three interventions. A p-value < 0.05 was considered significant. All enrolled participants were included in the adverse event reporting. This trial is registered with ClinicalTrials.gov, NCT04163874.

Between Feb 14, 2020, and Oct 5, 2021, 15 adults (9 females) and 17 adolescents (10 females) were enrolled; 2 female adolescents were excluded from primary outcome analyses for not completing all interventions. Compared to the insulin-and-placebo system with carbohydrate counting, non-inferiority in time in range was achieved with the insulin-and-pramlintide system and meal announcement (58% (SD 18) vs 63% (15); difference: 5% (12), p=0.031), but not with the insulin-and-placebo system and meal announcement (p=0.11). Mean Emotional Burden scores were not different between interventions. No serious adverse events were reported. 2 participants reported non-mild gastrointestinal symptoms on the insulin-and-pramlintide arm; no non-mild symptoms were reported on either other intervention.

Conclusion: The insulin-and-pramlintide with simple meal announcement intervention was non-inferior to the insulin-and-placebo with carbohydrate counting intervention. This novel dual hormone closed-loop system has the potential to alleviate carbohydrate counting with degrading glucose control.