An artificial pancreas that releases both insulin and pramlintide, an analog of amylin, might offer better control during the after-meal period.
Researchers have worked tirelessly over the past decade to create the so-called “artificial pancreas,” an easy-to-use medical device that can monitor glucose levels and deliver insulin to help individuals with Type 1 diabetes control their blood sugars. Last year, Medtronic, a medical device manufacturer, was the first to come to market with a closed-loop device that showed some success in helping individuals better manage their sugars. But while it was an important first step, Ahmad Haidar, a biomedical engineer specializing in control algorithms at Canada’s McGill University, says that there is room for improvement.
“These systems only deliver insulin to help lower glucose,” Haidar says. “And with the algorithms in the device, you can get control in a target range of about seventy to seventy five percent, significantly higher than traditional insulin pump devices. But that means you still have twenty five percent of your day, six or seven hours a day, on average, where you don’t have good control.”
The American Diabetes Association recommends that diabetics maintain blood glucose levels of 80-130 mg/dl before meals and under 180 mg/dl for one to two hours after meals. But it’s often easier said than done. One reason is when an individual with Type 1 diabetes sits down for a meal, food absorption often happens faster than insulin absorption, which results in that “high” after eating. Type 1 diabetics not only have a lack of insulin, their pancreases also don’t produce amylin, a hormone that works in concert with insulin to help regulate blood glucose levels. That got Haidar wondering if a device that released both insulin and Pramlintide, an analog of amylin, might offer better control during the after-meal period.
“If you eat, no matter how smart your device algorithm is, you’ll see sugar levels going high over the next few hours,” he says. “Pramlintide slows meal absorption so it is able to match the insulin absorption, so we offered the possibility that we could improve glucose control during those six hours that people are outside their target range.”
To test the idea, Haidar and his colleagues compared the current artificial pancreas system to one that released both insulin and pramlintide. The team regulated the pramlintide by using their own homegrown dosing algorithm designed to mimic coformulation (the packaging of more than one drug into a single treatment or pill) in 12 adults diagnosed with Type 1 diabetes over a 24-hour period.
They found the dual insulin-pramlintide system allowed the patients to stay in range about 85 percent of the time. The new system also showed lesser glucose level variability, or swings between glucose highs and lows, than the traditional artificial pancreas systems. The team presented its results at the American Diabetes Association’s 78th Scientific Sessions held this summer in Orlando, FL.
Haidar and his colleagues plan to follow this work with a small two-week at-home trial. While Haidar is pleased with the results, he believes that future artificial pancreas devices can offer even better glucose management for patients. It just takes time to find the right combination of hormones and/or medications as well as the appropriate control algorithms, he says.
“It may look like it has taken a long time to come up with the artificial pancreas, but if you compare it to drug development, it really isn’t that long,” Haidar says. “You see large variability between patients with Type 1 diabetes. It’s easy to develop one artificial pancreas for one person, but not as easy to develop one artificial pancreas that will work for hundreds of patients. Defining and testing these algorithms is important, so we can get to where we can offer significant improvements in glucose control without increasing patient burden. That’s our goal.”
Kayt Sukel is an independent technology writer.