You can read about the trial, conducted at Massachusetts General Hospital (MGH) in Boston, in the 14 April online issue of Science Translational Medicine.
Co-lead researcher Dr Steven Russell, of the MGH Diabetes Center, who is also an instructor in Medicine at Harvard Medical School, told the press that:
"This is the first study to test an artificial pancreas using both insulin and glucagon in people with type 1 diabetes."
"It showed that, by delivering both hormones in response to frequent blood sugar tests, it is possible to control blood sugar levels without hypoglycemia, even after high-carbohydrate meals," he added.
In the trial, Russell and the other co-lead investigator Dr Edward Damiano, an associate professor of Biomedical Engineering at Boston University, and colleagues, tested a new type of artificial pancreas developed at Boston Unversity.
The new artificial pancreas system comprises a blood glucose monitor, insulin pump technology and specialized software that controls the delivery of insulin and glucagon (a hormone that raises blood sugar levels).
In healthy people, blood sugar is controlled by a delicate balance between two hormones: insulin, which encourages various parts of the body to take up and use blood sugar, and glucagon, which raises blood sugar levels.
Both insulin and glucagon are produced in the pancreas: insulin by beta cells and glucagon by alpha cells.
However, in people with type 1 diabetes, their immune system wipes out the beta cells, so they don't have enough insulin, resulting in too much sugar in the blood (hyperglycemia).
To control their blood sugar level, type 1 diabetes patients have insulin treatment, which delays and can even prevent the long term effects of the disease, such as retinal damage, kidney failure and cardiovascular disease.
Insulin treatment for type 1 diabetes patients is further complicated by the fact that although their glucagon-producing alpha cells are intact, they don't respond to low levels of blood sugar, so if they get too much insulin, they are then at risk of developing a life-threatening drop in blood sugar, or hypoglycemia.
Damiano commented that:
"Large doses of glucagon are used as a rescue drug for people with severely low blood sugar."
Because of this elevated risk of hypoglycemia, the researchers at Boston University developed a new type of system that not only accounts for the rate of insulin absorption, but also includes glucagon.
"Our system is designed to counteract moderate drops in blood sugar with minute doses of glucagon spread out throughout the day, just as the body does in people without diabetes," explained Damiano.
The researchers tested their prototype system on diabetic pigs in 2007, as a result of which they got Food and Drug Administration (FDA) approval to test it on humans.
For the trial, Russell, Damiano and colleagues, recruited 11 adults with type 1 diabetes. The trial was designed primarily to test the software that controls the artificial pancreas.
To get the most accurate glucose readings they used a sensor placed directly into a vein instead of taking readings under the skin.
For 27 hours the system controlled the participants' glucose levels, during which time they had three standardized, high-carbohydrate meals and slept at the hospital overnight.
For six participants the system kept their glucose close to the target level, while five others had hypoglycemia that was sufficiently significant enough that they had to drink a dose of orange juice to raise their glucose level.
One of the results that surprised the researchers was they found large differences in insulin absorption rates among the patients, the fastest being four times faster than the slowest, which was much slower than expected. But they they were able to account for this by adjusting the system.
The software controlling the system was initially designed to dispense insulin at an expected absorption rate: this meant participants who absorbed at a slower rate got too much and developed hypoglycemia.
When they tested participants' response to a single insulin injection, the researchers verified that some had consistently slow and some had consistently faster insulin absorption rates.
They also observed that the rate of absorption varied a lot from experiment to experiment, even on an individual basis, to allow for individually specific dosage calculations.
So, they adjusted the software and globally lowered insulin absorption rate and repeated the experiments with the same participants.
The second time, none of the participants with slower absorption rates became sufficiently hypoglycemic to need intervention, and blood sugar levels were only slightly higher among those who had the faster insulin absorption rate.
This showed, said the researchers, that the adjusted software parameters worked for all the participants, and may be adequate for all type 1 diabetes patients.
They concluded that the elimination of hypoglycemic events in the second set of experiments in the same participants, confirmed that they were caused by a mismatch between the parameter settings in the software and the participants' absorption rates.
They pointed out that all previously published studies of artifical pancreas systems have reported hypoglycemic events, but this is the first time that one has confirmed and addressed their cause.
The researchers are now planning to follow up this study with another set of experiments that run for more than 48 hours, involving both adults and children. This time they will use the revised settings and an FDA-approved continuous glucose monitor.
They also plan to compare this two-hormone system with one that only uses insulin.
Russell said he imagined one day their system will be a wearable device, incorporating a glucose sensor that fits under the skin and communicates wirelessly with a pump about the size of a cell phone.
"The pump would administer insulin and probably glucagon, and would contain a microchip that runs the control software," he added.
Damiano has an 11 year old son who was diagnosed with type 1 diabetes when he was one year old. He said that such a solution would eliminate the need for patients to continually check their blood sugar and make treatment decisions every few hours.
He said while it would have to be maintained, it could take over the decision-making, closely emulating a functioning pancreas.
"It wouldn't be a cure, but it has the potential to be the ultimate evolution of insulin therapy for type 1 diabetes," said Damiano.
"A Bihormonal Closed-Loop Artificial Pancreas for Type 1 Diabetes."
Firas H. El-Khatib, Steven J. Russell, David M. Nathan, Robert G. Sutherlin, and Edward R. Damiano.
Sci Transl Med, Vol. 2, Issue 27, p. 27ra27, published online 14 April 2010.
Source: Massachusetts General Hospital.