The American Diabetes Association 71st Scientific Sessions:
Progress Reported in Clinical Trials of Artificial Pancreas
Clinical trials of artificial pancreases have moved into a critical new phase, lasting for as long as 2 weeks and including unannounced meals and exercise, according to studies presented at the 2011 American Diabetes Association meeting.
Each of the half dozen trials presented at a session at the meeting on June 26 involved modest numbers of patients, yet showed highly significant reductions in the number of both hypoglycemic and hyperglycemic episodes among type 1 diabetics while treated on the devices, compared to standard self-managed care.
The devices automatically release insulin based on readings from continuous glucose monitors.
Researchers were buoyed not only by the new studies, but by the presence at the session of Charles “Chip” Simliki, PhD, diabetes team leader in the US Food and Drug Administration’s (FDA) Center for Devices and Radiological Health. Days before the session, Dr. Simliki’s office released a long-awaited draft guidance describing what the FDA requires in order to approve a preliminary version of the artificial pancreas.
“The diabetes community needs these products,” said Dr. Simliki. The necessary study, he said, would likely take only about 3 months. “We’re not talking about a year-long study,” he said.
The first study presented (abstract 149-OR) was designed to test the ability of an automated sensor-controlled insulin pump to respond to the drop in insulin sensitivity caused by hydrocortisone. Researchers gave 8 patients on a so-called “closed-loop” system, in which a pump is automatically controlled by an algorithm based on sensor readings, 40 mg of hydrocortisone by mouth every 4 hours for a total of 7 times. By hour 28, the algorithm had adjusted to the decreased insulin sensitivity by increasing the amount of insulin delivered, resulting in a statistically significant improvement in blood-glucose levels.
The second study (150-OR) was designed to test the effects of a so-called “low glucose suspend” function, in which an insulin pump stops delivering insulin for 2 hours when a glucose sensor detects a low blood-glucose level. Comparing a Medtronic Paradigm Veo system against standard pump therapy in 21 children over 2 weeks, the study found significantly fewer blood glucose levels below 70 (1.27 per day vs 0.95 per day, p = 0.01), and the total time spent with hypoglycemia was nearly cut in half (101 minutes per day vs 58 minutes per day, p = 0.002).
A third study (154-OR), led by William V. Tamborlane, MD, of Yale University School of Medicine, studied the effects of daytime exercise on nighttime blood glucose levels while on either a closed-loop or open-loop system. During nights following daytime exercise, 72% of blood glucose levels were within target range (80-140 mg/dl), 8% were below, and 20% were above while on the closed-loop system, compared to 52% within target range, 21% below, and 27% above on the open-loop system (p = 0.006).
Dr. Simliki, who has type 1 diabetes but does not currently wear either a pump or a sensor, said he intends to use an artificial pancreas as soon as one is approved.
“I cannot wait until I can put one of these suckers on and not worry about my disease,” he said.