Fully Automated Closed Loop Control in Adolescents With Type 1 Diabetes

Maintaining blood glucose (BG) control among adolescents with Type 1 diabetes (T1D) is arguably the greatest challenge in the entire field of T1D. One reason for this poor control relates to missed meal boluses, which affects 65% of adolescents at least once weekly, with 38% missing at least 15% of their boluses. Adolescents who miss four boluses weekly experience an increase of 1% in their HbA1c. While the advent of the artificial pancreas (AP) offers promise of safe reductions in HbA1c, the study team previously found that the AP only partly compensates for missed prandial insulin -demonstrating that some form of meal announcement is necessary for good BG control, even with an AP. One way to automate this process is by sharing the prandial dosing responsibilities between an automated insulin priming (based on continuous glucose monitor condition predictive of a safe situation for such insulin dosing) and a closed loop controller capable of reconstructing (estimating) the prevailing glucose rate of appearance from an unannounced meal. The study team has developed such an insulin priming schema and integrated it into a new version of the robust Model Predictive Controller University of Virginia AP system (called the RocketAP).

In the current study, the investigators are testing this new AP system in two configurations: hybrid and fully automated, among up to 20 adolescents. The primary outcome will be one of efficacy in assessing how well the new system controls post-prandial BG in the absence of carbohydrate (CHO) announcement as compared to the same situation but using the Control-IQ closed loop algorithm, also designed at UVa and using the same modular architecture and safety system, but without insulin priming and with a less advanced model-based controller. Further comparisons will be made to BG control on RocketAP with CHO announcement and on Control-IQ with CHO announcement. Adolescents will be started on the respective University of Virginia AP systems (RocketAP and Control-IQ in random order, both implemented on the DiAs platform, MAF 2109) and followed over the course of two dinners on each of the two platforms: a dinner where CHO is announced as normal and the 2nd where no announcement is made.

The study team hypothesize that performances of RocketAP in fully automated mode will lie in between Hybrid and Fully Automated Control-IQ. In time, this may provide an opportunity to improve BG control among adolescents who miss meal announcement.

Inclusion criteria

• Age ≥12.0 and ≤25 years old at time of consent
• Clinical diagnosis, based on investigator assessment, of type 1 diabetes for at least one year
• Currently using insulin for at least six months
• Currently using insulin pump for at least three months
• Using insulin parameters such as carbohydrate ratio and correction factors consistently on their pump in order to dose insulin for meals or corrections
• Access to internet and willingness to upload data during the study as needed
• For females, not currently known to be pregnant or breastfeeding
• If female and sexually active, must agree to use a form of contraception to prevent pregnancy while a participant in the study. A negative serum or urine pregnancy test will be required for all females of childbearing potential. Participants who become pregnant will be discontinued from the study. Also, participants who during the study develop and express the intention to become pregnant within the timespan of the study will be discontinued.
• Willingness to suspend use of any personal CGM for the duration of the clinical trial once the study CGM is in use
• Willingness to use the UVa artificial pancreas system throughout study sessions.
• Willingness to switch to lispro (Humalog) or aspart (Novolog) if not using already, and to use no other insulin besides lispro (Humalog) or aspart (Novolog) during the study
• Total daily insulin dose (TDD) at least 10 U/day and not more than 100 U/d
• Willingness not to start any new non-insulin glucose-lowering agent during the course of the trial (including metformin, GLP-1 agonists, pramlintide, DPP-4 inhibitors, biguanides, sulfonylureas and naturaceuticals)
• Willingness to eat at least 1 g/kg of carbohydrate per day during the camp/hotel admission
• Willingness to reschedule Study Dinner Sessions if placed on oral steroids
• An understanding and willingness to follow the protocol and signed informed consent

Exclusion criteria

• History of diabetic ketoacidosis (DKA) in the 12 months prior to enrollment
• Severe hypoglycemia resulting in seizure or loss of consciousness in the 12 months prior to enrollment
• Pregnancy or intent to become pregnant during the trial
• Currently being treated for a seizure disorder
• Planned surgery during study duration
• Treatment with any non-insulin glucose-lowering agent (including metformin, GLP-1 agonists, pramlintide, DPP-4 inhibitors, SGLT-2 inhibitors, biguanides, sulfonylureas and naturaceuticals)
• A known medical condition that in the judgment of the investigator might interfere with the completion of the protocol.
• Use of an automated insulin delivery mechanism that is not downloadable by the subject or study team