Positive results from the first feasibility study of an advanced first-generation artificial pancreas system were presented at the 72nd Annual American Diabetes Association Meeting in Philadelphia. Findings from the study indicated that the Hypoglycemia-Hyperglycemia Minimizer (HHM) System was able to automatically predict a rise and fall in blood glucose and correspondingly increase and/or decrease insulin delivery safely. The HHM System included a continuous, subcutaneous insulin pump, a continuous glucose monitor (CGM) and special software used to predict changes in blood glucose. The study was conducted by Animas Corporation in collaboration with JDRF as part of an ongoing partnership to advance the development of a closed-loop artificial pancreas system for patients with type 1 diabetes. In June 2011, Animas received Investigational Device Exemption (IDE) approval from the U.S. FDA to proceed with human clinical feasibility studies for the development of a closed-loop artificial pancreas system. The company partnered with the JDRF in January 2010 to begin developing such an automated system to help people living with Type 1 diabetes better control their disease.
“We are encouraged by the results of the first human trials in this partnership with Animas,” said Aaron Kowalski, Ph.D., Assistant Vice President of Research at JDRF. “An artificial pancreas system that can not only detect, but can predict high and low blood sugar levels and make automatic adjustments to insulin delivery would be a major advance for people with type 1 diabetes. Such a system could alleviate a huge burden of managing this disease.”
The trial was a non-randomized, uncontrolled feasibility study of 13 participants with type 1 diabetes at one trial site in the United States. The investigational Hypoglycemia-Hyperglycemia Minimizer (HHM) system was studied for approximately 24 hours for each study participant during periods of open and closed loop control via a model predictive control algorithm with a safety module run from a laptop platform. Insulin and food variables were manipulated throughout the study time period to challenge and assess the system.
The primary endpoint was to evaluate the ability of the algorithm to predict a rise and fall in glucose above or below set thresholds and to command the pump to increase, decrease, suspend and/or resume insulin infusion accordingly. The secondary endpoint was to understand the HHM system’s ability to safely keep glucose levels within a target range and to provide guidance for future system development. The study also examined the relationship between CGM trends and the control model’s algorithm for insulin delivery.
