An investigational “artificial pancreas” system is the latest to improve glycemic control in patients with
type 1 diabetes in a real-world setting, this time for up to 6 months.
Findings from Protocol 3 of the National Institutes of Health (NIH) funded International Diabetes Closed Loop (
iDCL) trial of Tandem’s t:slim
insulin pump with Control-IQ technology were
published online October 16 in the
New England Journal of Medicine by Sue A. Brown, MD, of the University of Virginia, Charlottesville, and colleagues.
This is the first 6-month study of a closed-loop or artificial
pancreas system to include a dedicated control group, according to a
Tandem statement.
The closed-loop system, currently under review by the US Food and
Drug Administration (FDA), resulted in approximately 2.6 more hours per
day spent in the target blood glucose range of 70-180 mg/dL compared
with the control group, who used noncommunicating
insulin pumps and continuous glucose monitor (CGM) devices.
“Testing the safety and effectiveness of new technologies in
real-world settings is critical to prove the usability of these systems
by people with diabetes to achieve a better daily control of their blood
glucose levels,” study project scientist Guillermo Arreaza-Rubín, MD,
director of the Diabetes Technology Program, National Institute of
Diabetes and Digestive and Kidney Diseases (NIDDK), said in an NIH
statement.
“Impressive and Clinically Relevant,” but Caveats Remain
The artificial pancreas is an all-in-one system that tracks blood
glucose using a CGM and automatically delivers insulin when needed using
an insulin pump. It replaces reliance on blood glucose testing by
fingerstick or CGM with separate delivery of insulin by pump or multiple
daily injections.
One such system, the Medtronic
MiniMed 670G artificial pancreas, is already approved by the FDA for use in those age 7 years and older with type 1 diabetes.
These latest iDCL study results are “impressive and clinically
relevant,” says Daniela Bruttomesso, MD, PhD, University of Padua,
Italy, in an
accompanying editorial.
This is because it has “been shown that for each 10% reduction in the
time spent in the glucose target range, the risk of development or
progression of retinopathy increases by 64% and the risk of development
of microalbuminuria by 40%,” she observed.
However, Bruttomesso cautions — as do the authors — that the trial
was performed at university-based diabetes centers and the baseline use
of sensors and insulin pumps by participants was higher than in the
general type 1 diabetes population.
Moreover, the sensor-augmented pump system used in the control group
lacked the ability to suspend the pump’s insulin delivery for an
impending low glucose, a feature of the currently available Medtronic
MiniMed 670G.
Commenting on the first generation of artificial pancreas systems in
general, Bruttomesso observes: “We are not there yet, but [the current
study provides] a big step toward a brighter future for patients.”
“Time in Range,” Particularly Overnight, Improved
The new study randomized 168 individuals age 14 years and older (mean
33 years) with type 1 diabetes to the closed-loop system (112 patients)
or a sensor-augmented insulin pump (56 patients).
The closed-loop system combines a
t:slim X2 insulin pump and a Dexcom G6 CGM, along with an algorithm with a dedicated
hypoglycemia
safety module, automated boluses to correct high blood sugars, and
overnight intensification of basal insulin delivery aimed at normalizing
morning glucose values.
The system predicts glucose levels 30 minutes ahead and adjusts
insulin delivery to help prevent highs and lows, but users must still
manually bolus for meals.
At baseline, 79% of patients were already using pumps and 70% were using CGMs.
For the primary endpoint, mean time in target range, the closed-loop
group increased from 61% at baseline to 71% by 1 month, and this
remained consistent at 6 months, while the time in range percentage
remained at 59% in the control group across the study period.
The 11 percentage-point difference (
P < .001) corresponded to 2.6 more hours in range per day in the closed-loop group, Brown and colleagues report.
The difference in percentage time-in-range in the closed-loop and
control groups was particularly pronounced from midnight to 6:00 am (76%
vs 59%), reflecting the “increased aggressiveness of the algorithm to
meet a lower glucose target during the second half of the night,” the
authors explain.
The overall mean percentage of time spent with glucose levels >
180 mg/dL at 6 months in the closed-loop versus control group was 27%
versus 38% (
P < .001), corresponding to a mean difference of 2.4 hours per day.
Hemoglobin A1c values were also significantly lower at 6 months in the closed-loop versus control group (7.06% vs 7.39%;
P = .001).
Percentage of time with glucose levels < 70 mg/dL were 1.58% and 2.25%, respectively (
P
< .001), amounting to 13 minutes per day. Glucose levels < 54
mg/dL were rare in both groups but still significantly lower in the
closed-loop group (0.29% vs 0.35%;
P = .02).
Severe hypoglycemia (requiring assistance) didn’t occur in either group. One participant in the closed-loop group experienced
diabetic ketoacidosis resulting from a pump infusion set failure.
Barriers Remain to a Fully Automated Closed-Loop System
“This research shows that this artificial pancreas system has the
potential to improve the health of people living with type 1 diabetes,
while also potentially lifting much of the burden of care from those
with the disease and their caregivers,” notes Arreaza-Rubín.
However, Bruttomesso
points out that first-generation closed-loop systems are not without their teething problems.
Although “the closed-loop system is becoming a mature technology
ready for practical use…there are a variety of barriers to a fully
automated closed-loop system,” including slow subcutaneous absorption of
insulin, insufficient sensor accuracy, and algorithms that fall short
of accounting for everyday insulin needs, she points out.
Indeed, one study
reported earlier this year
showed that over a third of patients using the currently available
Medtronic MiniMed 670G quit using the system within 9 months because of
technical difficulties.
And Bruttomesso adds, “Whether closed-loop systems can be used in
higher-risk patients, such as those with impaired awareness of
hypoglycemia, also remains a pressing issue.”
“Cost-effectiveness, user acceptance, and training of both patients
and healthcare professionals also need to be addressed,” she emphasizes.
“It is clear that patients would appreciate wearing devices that
require minimal interaction, leading to a more carefree lifestyle,” she
concludes.
The study was supported by a grant from the NIDDK and University
of Virginia. Brown has reported receiving grant support and supplies,
paid to her institution, from Tandem Diabetes Care, Dexcom, and Roche
Diagnostics. Bruttomesso has reported no relevant financial relationships.
N Engl J Med. Published online October 16, 2019.
Abstract,
Editorial
https://www.medscape.com/viewarticle/919986#vp_1