User login
MUNICH – A “bionic pancreas” that delivers glucagon as well as insulin fared well against conventional insulin pump therapy, significantly reducing mean glucose levels and minimizing time spent in hypoglycemia.
The iLet bionic pancreas is being developed by Beta Bionics in conjunction with Eli Lilly and with support from the National Institutes of Health. Beta Bionics bills itself as a public benefit corporation – “a for-profit entity also dedicated to social responsibility and a public benefit mission,” according to an article published by Boston University.
Glucagon is the key that takes the bionic pancreas above and beyond what current insulin pump systems can do, said Dr. Russell, an endocrinologist at Massachusetts General Hospital, Boston.
“Even the pancreas, which has all the advantages of releasing insulin right into the portal vein and directly sensing glucose in the blood, uses the countering hormone, glucagon, to prevent hypoglycemia, particularly during exercise and in the late postprandial phase,” he said. “We are trying to mimic that capability. Glucagon gives us an additional tool to further reduce the risk of hypoglycemia.”
The algorithm is almost completely independent of user input – another key benefit, Dr. Russell said. It initializes with input about the patient’s weight and adapts its insulin delivery by essentially learning each user. There is no need to count carbohydrates, for example. The system “learns” over time its user’s typical meal patterns and can be programmed to deliver insulin accordingly.
“The user can enter, for example, ‘typical lunch,’ and the system will dispense some insulin before the meal and the rest later, in automatic delivery mode.”
The iLet has been studied in several settings and populations including, most recently, a successful crossover trial in 19 children at a diabetes camp (Lancet Diabetes Endocrinol. 2016; 4[3]:233-43).
The device was similarly successful in both teens and adults.
The study Dr. Russell reported at EASD 2016 comprised 39 patients with type 1 diabetes who were using an insulin pump. The mean age of the patients was 33 years, and mean disease duration was 17 years. The mean daily insulin dose was 0.6 U/kg. The mean baseline hemoglobin A1c was 7.7% and mean blood glucose, 176 mg/dL.
The crossover trial randomized patients to 11 days of treatment with their existing Medtronic insulin pump or the iLet system. The primary outcomes were the mean glucose level and time spent in hypoglycemic range (less than 60 mg/dL).
Under the usual care arm, “We saw a wide range of glucose values, which became much tighter when patients were using the bionic pancreas,” Dr. Russell said.
The overall average glucose level was 9 mmol/L in the usual care arm vs. 7.8 mmol/L in the bionic pancreas arm. Time in hypoglycemia was significantly reduced (27 minutes/24 hours vs. 9 minutes/24 hours).
“The standard deviation of the mean glucose was also larger in usual care, which is consistent with the automatic adaptive function of the bionic pancreas. If the mean glucose gets too high, it treats more aggressively; and if glucose is too low, it becomes less aggressive.”
The average amount of insulin used per day was similar (0.62 U/kg vs. 0.66 U/kg) and the average amount of glucagon used was 0.5 mg/day in each group.
During the night, the bionic pancreas kept mean blood glucose lower (134 mg/dL vs. 165 mg/dL) and more stable, reducing time in hypoglycemia (1.4 minutes/night vs. 4.8 minutes/night).
There was one incident of severe hypoglycemia in the usual-care arm, and none in the bionic-pancreas arm. There was a statistically significant increase in nausea associated with the glucagon, Dr. Russell noted, but the impact was still quite small. On a 0-10 rating scale, nausea in the bionic pancreas group was rated a 0.5 compared to a 0.05 in the usual care arm.
There were no changes in blood pressure, weight, or any lab test.
The ongoing studies continue to show “that automated bihormonal control of glycemia in the home-use setting is feasible,” Dr. Russell said. “Micro-dose glucagon was well tolerated and the bihormonal pancreas reduced both mean glucose and hypoglycemia relative to the patients’ usual care devices.”
The iLet continues to undergo modifications that are making it more user friendly, he added.
Dr. Stevens disclosed that he has a patent pending on “certain aspects” of the device, and that he is an unpaid scientific consultant for Beta Bionics. He also disclosed financial relationships with several pharmaceutical and device companies, including Eli Lilly and Medtronic.
MUNICH – A “bionic pancreas” that delivers glucagon as well as insulin fared well against conventional insulin pump therapy, significantly reducing mean glucose levels and minimizing time spent in hypoglycemia.
The iLet bionic pancreas is being developed by Beta Bionics in conjunction with Eli Lilly and with support from the National Institutes of Health. Beta Bionics bills itself as a public benefit corporation – “a for-profit entity also dedicated to social responsibility and a public benefit mission,” according to an article published by Boston University.
Glucagon is the key that takes the bionic pancreas above and beyond what current insulin pump systems can do, said Dr. Russell, an endocrinologist at Massachusetts General Hospital, Boston.
“Even the pancreas, which has all the advantages of releasing insulin right into the portal vein and directly sensing glucose in the blood, uses the countering hormone, glucagon, to prevent hypoglycemia, particularly during exercise and in the late postprandial phase,” he said. “We are trying to mimic that capability. Glucagon gives us an additional tool to further reduce the risk of hypoglycemia.”
The algorithm is almost completely independent of user input – another key benefit, Dr. Russell said. It initializes with input about the patient’s weight and adapts its insulin delivery by essentially learning each user. There is no need to count carbohydrates, for example. The system “learns” over time its user’s typical meal patterns and can be programmed to deliver insulin accordingly.
“The user can enter, for example, ‘typical lunch,’ and the system will dispense some insulin before the meal and the rest later, in automatic delivery mode.”
The iLet has been studied in several settings and populations including, most recently, a successful crossover trial in 19 children at a diabetes camp (Lancet Diabetes Endocrinol. 2016; 4[3]:233-43).
The device was similarly successful in both teens and adults.
The study Dr. Russell reported at EASD 2016 comprised 39 patients with type 1 diabetes who were using an insulin pump. The mean age of the patients was 33 years, and mean disease duration was 17 years. The mean daily insulin dose was 0.6 U/kg. The mean baseline hemoglobin A1c was 7.7% and mean blood glucose, 176 mg/dL.
The crossover trial randomized patients to 11 days of treatment with their existing Medtronic insulin pump or the iLet system. The primary outcomes were the mean glucose level and time spent in hypoglycemic range (less than 60 mg/dL).
Under the usual care arm, “We saw a wide range of glucose values, which became much tighter when patients were using the bionic pancreas,” Dr. Russell said.
The overall average glucose level was 9 mmol/L in the usual care arm vs. 7.8 mmol/L in the bionic pancreas arm. Time in hypoglycemia was significantly reduced (27 minutes/24 hours vs. 9 minutes/24 hours).
“The standard deviation of the mean glucose was also larger in usual care, which is consistent with the automatic adaptive function of the bionic pancreas. If the mean glucose gets too high, it treats more aggressively; and if glucose is too low, it becomes less aggressive.”
The average amount of insulin used per day was similar (0.62 U/kg vs. 0.66 U/kg) and the average amount of glucagon used was 0.5 mg/day in each group.
During the night, the bionic pancreas kept mean blood glucose lower (134 mg/dL vs. 165 mg/dL) and more stable, reducing time in hypoglycemia (1.4 minutes/night vs. 4.8 minutes/night).
There was one incident of severe hypoglycemia in the usual-care arm, and none in the bionic-pancreas arm. There was a statistically significant increase in nausea associated with the glucagon, Dr. Russell noted, but the impact was still quite small. On a 0-10 rating scale, nausea in the bionic pancreas group was rated a 0.5 compared to a 0.05 in the usual care arm.
There were no changes in blood pressure, weight, or any lab test.
The ongoing studies continue to show “that automated bihormonal control of glycemia in the home-use setting is feasible,” Dr. Russell said. “Micro-dose glucagon was well tolerated and the bihormonal pancreas reduced both mean glucose and hypoglycemia relative to the patients’ usual care devices.”
The iLet continues to undergo modifications that are making it more user friendly, he added.
Dr. Stevens disclosed that he has a patent pending on “certain aspects” of the device, and that he is an unpaid scientific consultant for Beta Bionics. He also disclosed financial relationships with several pharmaceutical and device companies, including Eli Lilly and Medtronic.
MUNICH – A “bionic pancreas” that delivers glucagon as well as insulin fared well against conventional insulin pump therapy, significantly reducing mean glucose levels and minimizing time spent in hypoglycemia.
The iLet bionic pancreas is being developed by Beta Bionics in conjunction with Eli Lilly and with support from the National Institutes of Health. Beta Bionics bills itself as a public benefit corporation – “a for-profit entity also dedicated to social responsibility and a public benefit mission,” according to an article published by Boston University.
Glucagon is the key that takes the bionic pancreas above and beyond what current insulin pump systems can do, said Dr. Russell, an endocrinologist at Massachusetts General Hospital, Boston.
“Even the pancreas, which has all the advantages of releasing insulin right into the portal vein and directly sensing glucose in the blood, uses the countering hormone, glucagon, to prevent hypoglycemia, particularly during exercise and in the late postprandial phase,” he said. “We are trying to mimic that capability. Glucagon gives us an additional tool to further reduce the risk of hypoglycemia.”
The algorithm is almost completely independent of user input – another key benefit, Dr. Russell said. It initializes with input about the patient’s weight and adapts its insulin delivery by essentially learning each user. There is no need to count carbohydrates, for example. The system “learns” over time its user’s typical meal patterns and can be programmed to deliver insulin accordingly.
“The user can enter, for example, ‘typical lunch,’ and the system will dispense some insulin before the meal and the rest later, in automatic delivery mode.”
The iLet has been studied in several settings and populations including, most recently, a successful crossover trial in 19 children at a diabetes camp (Lancet Diabetes Endocrinol. 2016; 4[3]:233-43).
The device was similarly successful in both teens and adults.
The study Dr. Russell reported at EASD 2016 comprised 39 patients with type 1 diabetes who were using an insulin pump. The mean age of the patients was 33 years, and mean disease duration was 17 years. The mean daily insulin dose was 0.6 U/kg. The mean baseline hemoglobin A1c was 7.7% and mean blood glucose, 176 mg/dL.
The crossover trial randomized patients to 11 days of treatment with their existing Medtronic insulin pump or the iLet system. The primary outcomes were the mean glucose level and time spent in hypoglycemic range (less than 60 mg/dL).
Under the usual care arm, “We saw a wide range of glucose values, which became much tighter when patients were using the bionic pancreas,” Dr. Russell said.
The overall average glucose level was 9 mmol/L in the usual care arm vs. 7.8 mmol/L in the bionic pancreas arm. Time in hypoglycemia was significantly reduced (27 minutes/24 hours vs. 9 minutes/24 hours).
“The standard deviation of the mean glucose was also larger in usual care, which is consistent with the automatic adaptive function of the bionic pancreas. If the mean glucose gets too high, it treats more aggressively; and if glucose is too low, it becomes less aggressive.”
The average amount of insulin used per day was similar (0.62 U/kg vs. 0.66 U/kg) and the average amount of glucagon used was 0.5 mg/day in each group.
During the night, the bionic pancreas kept mean blood glucose lower (134 mg/dL vs. 165 mg/dL) and more stable, reducing time in hypoglycemia (1.4 minutes/night vs. 4.8 minutes/night).
There was one incident of severe hypoglycemia in the usual-care arm, and none in the bionic-pancreas arm. There was a statistically significant increase in nausea associated with the glucagon, Dr. Russell noted, but the impact was still quite small. On a 0-10 rating scale, nausea in the bionic pancreas group was rated a 0.5 compared to a 0.05 in the usual care arm.
There were no changes in blood pressure, weight, or any lab test.
The ongoing studies continue to show “that automated bihormonal control of glycemia in the home-use setting is feasible,” Dr. Russell said. “Micro-dose glucagon was well tolerated and the bihormonal pancreas reduced both mean glucose and hypoglycemia relative to the patients’ usual care devices.”
The iLet continues to undergo modifications that are making it more user friendly, he added.
Dr. Stevens disclosed that he has a patent pending on “certain aspects” of the device, and that he is an unpaid scientific consultant for Beta Bionics. He also disclosed financial relationships with several pharmaceutical and device companies, including Eli Lilly and Medtronic.
Key clinical point:
Major finding: The overall average glucose level was 9 mmol/L in the usual care arm vs. 7.8 mmol/L in the bionic pancreas arm.
Data source: The randomized crossover trial comprised 39 patients with type 1 diabetes.
Disclosures: Dr. Stevens disclosed that he has a patent pending on “certain aspects” of the device and that he is an unpaid scientific consultant for Beta Bionics. He disclosed financial relationships with several pharmaceutical and device companies, including Eli Lilly and Medtronic.