Trial Backs CGM-Guided Insulin Dosing in Type 1 Diabetes

Miriam E Tucker

February 17, 2017

Use of a continuous glucose monitoring (CGM) system without confirmatory finger-stick checks is safe and effective in adults with well-controlled type 1 diabetes who are at low risk for severe hypoglycemia, new trial results show.

Findings from the Randomized Trial Comparing Continuous Glucose Monitoring With and Without Routine Blood Glucose Monitoring in Adults With Type 1 Diabetes (REPLACE-BG) were published online February 16 in Diabetes Care by Grazia Aleppo, MD, of Northwestern University, Chicago, Illinois, and colleagues.

In December 2016, the US Food and Drug Administration approved a label change for the Dexcom G5 CGM system allowing patients on pumps or multiple daily injections to use its readings to replace finger-stick monitoring in determining insulin doses.

Prior to that, the devices were licensed only as "adjunctive," and confirmatory finger sticks were still required by the label, although many patients were already known to be skipping them and relying solely on the CGM's readings.

With the current labeling, twice-daily finger sticks are still required for calibrating the system and for other circumstances such as when patient's symptoms don't match the readings or when the sensor's signal is lost.

The FDA's ruling followed a July 2016 hearing where Dexcom representatives presented data from two 1-week studies of 130 adults and children aged 2 years and older in which the Dexcom readings were compared with both a glucose meter and a laboratory reference.

REPLACE-BG, which the researchers presented today at International Conference on Advanced Technologies & Treatments for Diabetes in Paris, France, was said to be the first randomized trial to compare the safety and effectiveness of CGM with vs without confirmatory finger-stick blood glucose monitoring (BGM) to make therapeutic decisions in people with type 1 diabetes.

"The study results are supportive of the FDA decision. Based on the study results, clinicians and patients should have a reasonable comfort level for dosing of insulin based on CGM alone," senior author Roy W Beck, MD, PhD, executive director of the Jaeb Center for Health Research, Tampa, Florida, told Medscape Medical News.

However, Dr Beck added, "It is important that the sensor be properly calibrated. The Dexcom sensor used in the trial requires two blood glucose measurements per day for calibration. These need to be done properly — for instance, following hand washing, with a good blood glucose meter to ensure CGM sensor accuracy."

Asked to comment, David T Ahn, MD, an endocrinologist and diabetes technology expert at the University of California, Los Angeles, told Medscape Medical News, "I think it's great news," pointing out that one major benefit of the FDA label change was that it led to the long-awaited Medicare decision to reclassify CGMs as reimbursable durable medical equipment.

And he added that the decision also further paves the way for progress toward developing closed-loop insulin delivery systems, of which CGMs are primary components. "I'm a big believer in that."

However, Dr Ahn cautioned that clinicians still need to ensure that patients keep glucose test strips and conventional meters on hand, for several reasons in addition to the twice-daily calibration and for when the sensor fails. "I tell patients to use common sense to be sure the numbers seem reasonable, and if there's any concern to double-check with the meter."

And, Dr Ahn noted, it has been noted anecdotally the sensor's accuracy tends to be somewhat lower on the first day of use compared with subsequent days.

Overall, he said, "I think it's great that you have that option to trust your [CGM], but you're not bound to that as your only tool."

Noninferiority Demonstrated for Efficacy, Safety

The 26-week trial was conducted at 14 sites with 226 adults with type 1 diabetes, all of whom used insulin pumps and were in relatively good glycemic control (mean HbA1c 7.0%). Just under half (47%) were already using CGMs.

They were randomized 2:1 to CGM-only or CGM plus BGM. The CGM-only group was instructed to base all management decisions on the CGM readings, including adjustments for up or down trend arrows, with certain exceptions, including within the first 12 hours after the weekly insertion of a new sensor, on sick days, after taking acetaminophen (which interferes with CGM function), and when symptoms don't match the readings.

The CGM+BGM group was instructed to perform a BGM measurement whenever an insulin bolus was given, when managing hypoglycemia, and before going to sleep. Both groups performed BGM measurements for the twice-daily calibrations, and whenever readings were above 300 mg/dL (along with ketone tests).

A total of 142 CGM-only and 75 BGM+CGM patients completed the trial. The mean time spent with blood glucose levels 70 to 180 mg/dL at 26 weeks was 63% for the CGM-only group and 65% with CGM+BGM, meeting the prespecified noninferiority limit of 7.5%.

There were few changes from baseline to 26 weeks and no significant differences between groups in any metrics of glucose control, including mean glucose, hyperglycemia, hypoglycemia, or glycemic variability. Results were also similar in subgroups based on age, duration, education, CGM use prior to the study, baseline HbA1c, and baseline time in range.

No severe hypoglycemic events occurred in the CGM-only group, while one occurred in the CGM+BGM group. No patient in either group experienced diabetic ketoacidosis, although a blood ketone level of 0.6 mmol/L or greater occurred at least once in 32% of the CGM-only group and in 34% of the CGM+BGM group, but the difference wasn't significantly different (P = .79). Ketone levels of 1.0 mmol/L or greater also occurred at similar rates, 18% and 19% (P = .84).

There were also no differences between the groups in scores on the Hypoglycemia Fear Survey at 26 weeks (32 vs 31, P = .88).

Will the Results Hold in the Real World?

The authors acknowledge that the "major limitation" of the trial relates to generalizability of the results, given that the T1D Exchange population is seen at specialty clinics and all subjects wore insulin pumps and were in relatively good glycemic control without excessive hypoglycemia or hypoglycemic unawareness.

Nonetheless, Dr Beck told Medscape Medical News, "I believe that it is likely that the same degree of safety is present in injection users using CGM who are under good control, as there is no reason to expect different results. But we can't directly conclude this from the study. The safety also should be the same for both pump users and injection users who are not under good control, provided that they properly calibrate the sensor."

And he pointed out that in a survey of 999 T1D Exchange patients using CGM prior to the FDA label change, only 26% indicated that they always confirmed the CGM glucose concentration with a BGM measurement before giving an insulin bolus, and 41% admitted that they did so less than half the time. Yet, Dr Beck said, there was no indication of a difference in severe hypoglycemia between those who confirmed the premeal reading most or all of the time and those who didn't.

Another possible real-world consideration — again anecdotally — is that some patients will "push" their sensors beyond the FDA-approved 7 days by stopping and restarting the system without replacing the sensor, either for convenience or financial reasons. Data for accuracy beyond 7 days have not been available.

Asked about that, Dr Ahn said "I think that's a very valid concern, but that's why the FDA labeling is the way it is. When people extend beyond that, they're taking a calculated risk. With any product, when you go outside of the labeling, you're taking on some degree of risk."

Dr Ahn will often advise patients just starting out with CGM to use finger sticks frequently at first in order to "build trust" with the CGM. "Some patients don't trust it. But for those who do, it's great that with this new FDA labeling you can trust it....Bottom line is it really warrants the user to be very proactive and to take ownership of the data."

The study was funded by the Leona M and Harry B Helmsley Charitable Trust, with the CGM systems provided by Dexcom. Dr Aleppo has served on a scientific advisory board for Diasend and Novo Nordisk and received consultancy payments from Dexcom. Her employer has received research support from Novo Nordisk and Bristol-Meyers Squibb/AstraZeneca with no personal income to her. Dr Beck's employer has received research support and study supplies from Dexcom and Abbott Diabetes Care. Disclosures for the coauthors are listed in the paper. Dr Ahn has no relevant financial relationships.

For more diabetes and endocrinology news, follow us on Twitter and on Facebook.

Diabetes Care. Published online February 16, 2017. Abstract

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....