The world’s first artificial pancreas is coming to market much sooner than anyone expected – including the University of Colorado School of Medicine researchers whose work has been so central to the device’s development and approval.
The U.S. Food and Drug Administration on Sept. 28 approved the Medtronic MiniMed 670G system for patients with type 1 diabetes of ages 14 and older. The approval cites a clinical trial for which CU’s Barbara Davis Center for Childhood Diabetes was the largest proving ground.
“It’s a game-changer for people with diabetes,” said Satish Garg, MD, chief of the Barbara Davis Center’s adult clinic, who led the local study cohort.
Garg’s colleague Robert Slover, MD, who directs the Barbara Davis Center’s pediatric clinic, led the study’s largest pediatric site. He called the approval “historic.”
“We’ve been working on this for the last 15 or 20 years one way or another – all the way from pump development and sensor development to algorithm development,” Slover said. “I think it will have enormous impact.”
Three in one
The artificial pancreas is in fact a merging of three devices: a continuous glucose monitor (CGM) and an insulin pump, plus a glucose meter to calibrate the CGM. If blood glucose rises past a certain point, the CGM tells the pump to deliver more insulin to bring it back down; if blood sugar drops – or if algorithms in the CGM developed with the help of Barbara Davis Center researchers predict that it’s about to drop – the CGM tells the pump to stop infusing insulin to give the patient’s blood glucose a better chance to recover (technically, it’s called “predicted low-glucose suspend”). A smartphone-based app serves as a monitoring station for patients, parents or caregivers.
The speed of approval – Garg said the FDA took just three months to make a decision that typically waits a year – took even Medtronic by surprise. As a result, the device is first expected to be available in spring 2017.
The closed-loop pancreas has been a long time coming, with the Barbara Davis Center playing a big role throughout. The center’s expertise in diabetes care, its large patient base and its skills and infrastructure in supporting clinical trials has made it a favored partner of technologists in academia and industry who are developing artificial pancreas hardware and software.
“This is going to make diabetes care so much easier,” Chase said. “It’s exciting for our kids. It’s exciting for our families.”
The artificial pancreas does not cure type 1 diabetes, an autoimmune disease that about a million U.S. adults and 200,000 kids have. With type 1 diabetes, one’s own immune system kills the pancreatic islet cells that produce insulin, a protein that helps move glucose – a key metabolic fuel – from the bloodstream to the cells. The Medtronic device and others expected to follow in the coming years take a big step toward automating the difficult and error-prone process of controlling blood sugar levels, thereby minimizing the long-term eye, kidney, nerve and cardiovascular damage diabetes can cause.
The data show Medtronic’s artificial pancreas to work better than insulin pumps whose delivery depends solely on human intervention. The key statistic, Chase says, is that the Medtronic device kept the trial’s participants within the desired blood glucose range about 75 percent of the time.
“With current methods, we aim for 50 percent, and it’s rare that we get that,” Chase said. “It’s going to bring [blood glucose levels] into almost the normal range, and it’s going to greatly reduce the likelihood of microvascular complications. In five years, I would predict that eye, kidney, and nerve damage will be reduced by at least half.”
Slover added that the device’s guardrail against hypoglycemia – crashing blood sugar – will let providers be more aggressive with insulin dosing and improve average blood glucose levels, something doctors have “wanted to do for years.”
The device’s greatest strength is ability to control glucose levels when patients can’t. Nighttime control has been a huge problem: among adults, more than half of diabetic seizures from plunging blood sugar levels happen during sleep. Among children, it’s higher yet – 75 percent. Maintaining nighttime control was a major artificial pancreas goal from the beginning, Chase said. CGMs fitted with low-glucose alarms can help, but his group, with an assist from a military night-vision camera, found that 71 percent of teenagers sleep right through them.
Slover’s pediatric trial showed the Medtronic device to be safe and effective for kids as young as eight years old, and larger trials with this age group should soon lead to FDA approval. The system has yet to be tested in children under the age of eight, but those studies are also planned. Meanwhile, use in children under the age of 14 constitutes “off-label” use, and use by children under the age of eight is untested.
“But that research will get done and then the age range will get extended,” Slover said.
If Garg’s study participants were any measure, the artificial pancreas should fly off the shelves – assuming insurance coverage for a device estimated to cost several thousand dollars. Eighty-five percent of the Garg’s adult-study participants asked if they could keep the device after the three-month study ended, which Medtronic and the FDA let them do. Slover said the Barbara Davis Center expects to run 25 education and training sessions a week once the device is on the market. He says he and colleagues are getting emails from patients all over the world.
“A family in Saudi Arabia wanted to know when they could come and get their pump,” Slover said.
Michael McDermott, MD, medical director of University of Colorado Hospital’s endocrinology, metabolism and diabetes practice, said he would strongly recommend the device.
“I think this is the direction that type 1 diabetes management needs to go, and I hope someday that people with type 2 diabetes will have the same coverage,” McDermott said.
McDermott said about half of UCH’s roughly 1,500 type 1 diabetes patients are already using an insulin pump and/or CGM, which would make for a straightforward transition. But he added that some patients who still prefer to do manual finger-pricks and injections may also make the switch.
“They’ve been waiting for a big quantum leap as opposed to the incremental advances,” McDermott said.
But he cautioned that the Medtronic artificial pancreas is still a long way from being as sophisticated as the actual bodily organ. The data show the device to work well between meals, but the wearer still must estimate carbohydrates and enter them into the pump before meals so the device can deliver a pre-meal insulin booster, which takes about 20 minutes to take effect (the pancreas does the same thing with no delay because it delivers insulin directly into the digestive tract). Patients must also calibrate the CGM a couple of times a day with a drop of finger-prick blood.
Future devices will probably enter meals into the equation, McDermott said. Chase said he also anticipates later versions to be attuned to individual traits and behaviors, so as to adapt quickly to a young child who’s insulin-sensitive or a senior who’s insulin-resistant. Other improvements might include recognizing regular exercise patterns, say for a football player who practices a certain time every day – or even illness, which also affects metabolism. Also, future devices might deliver not only insulin, but also glucagon, which the pancreas secretes to counteract low blood sugar, and perhaps also incretins that help regulate the gut’s sugar absorption.
“Later generations will have a much finer ability to make adjustments for individual sensibilities,” Chase said.
Garg’s team is already testing Medtronic’s next-generation device, he says. Other device makers are at least a year behind, Chase estimates, but he expects the first competitors’ offerings to hit the market in late 2017 or 2018.
Chase wonders if the surprisingly rapid success of the artificial pancreas will refocus the attention of major funders such as the National Institutes of Health and JDRF – the latter which has spent $116 million on artificial pancreas projects over the past decade – on curing type 1 diabetes.
“We’re making inroads with immunosuppression and islet transplantation,” Chase said. “We’ve got a long ways to go yet, but there’s the possibility for a cure in the future.”
For now, Slover is enjoying the moment. He, Garg and Chase have been around long enough to remember the days before reliable blood glucose testing or fast-acting insulin, he said. Then came the first insulin pumps, which were “big like bricks.”
“And to think that we’ve moved from there to a technology that really will protect kids from hypo- and hyperglycemia most of the time,” Slover said. “It’s just amazing and wonderful.’’