Artificial Pancreas Control System in an Outpatient Setting

The objectives of this outpatient research study are (1) to assess the ability of this automated system to be operated by a subject with limited professional oversight; (2) to assess whether the new devices (Dexcom Gen 4 sensors, Motorola ES400 smart phone, iDex pump controller) will reduce the frequency of hardware and data communication lapses seen in the previous system; and (3) to measure the degree of glucose control achievable with this automated system. The system will adjust blood glucose by administering insulin and glucagon. Insulin is a hormone that lowers blood glucose and will be given nearly continuously during this study. Glucagon raises blood glucose and will be automatically administered during hypoglycemia. Both are natural hormones made by people without diabetes. Each subject will have four devices placed on his abdomen: two Omnipod insulin pumps, one for delivering insulin and one for delivering glucagon, and two Dexcom G4 glucose sensors for measuring glucose. The two sensors will feed glucose data into Motorola smart phone master controller, which will calculate the correct amount of insulin or glucagon to deliver. The system will then send the command to the correct Omnipod through the iDex pod controller. In this new system, the research subject will be able to monitor the progress of the study by use of the smart phone graphical user interface. The subject will have a companion with him/her during the entire study for safety purposes. Both the subject and companion will complete a training course on how to treat diabetic emergencies and how to operate the system. A study physician and technician will be in the hotel during each study and will be monitoring the study via a cloud-based data communication system. These studies will be carried out in a hotel setting.

The objective of the current human study is to verify the components of the Artificial Pancreas Control Software during an outpatient study. This master controller software is designed to be used in conjunction with two subcutaneous continuous glucose monitoring systems to regulate blood glucose levels as well as two Omnipod pumps, one for administering insulin and one for administering glucagon. The sensors communicate wirelessly with two sensor receivers which will be interfaced with the APC by wireless USB connection. The insulin and glucagon pumps will be controlled by the APC through a wireless USB connection. The algorithm included in the APC is an automated version of the Adaptive Proportional Derivative (APD) insulin and glucagon control algorithm, which was previously studied as an investigational device. The Automated APD has been studied in vivo (in 13 experiments, each 28 hr in length, with automated adjustment of pumps) and no serious adverse effects were noted. The APC will be tested in vivo during 28 hour experiments in an outpatient setting in preparation for home testing.