A next-generation implanted continuous glucose monitor (Eversense, Senseonics/Ascensia) is accurate and safe for up to 6 months of use, new data suggest.
The first-generation 90-day Eversense was approved by the US Food and Drug Administration in June 2018 and is currently the only long-term implanted continuous glucose monitor (CGM) available in the United States and Europe.
An older 180-day version (Eversense XL) has been available in Europe since 2017. The new second-generation 180-day system is currently under regulatory review in both the United States and Europe.
The Eversense consists of a fluorescence-based sensor, a transmitter, and a smartphone app that displays glucose values, trends, and alerts. The sensor is inserted subcutaneously into the upper arm by a certified healthcare professional in a brief office procedure. The transmitter is placed on the skin on top of the sensor.
The system includes an on-body vibratory alert as well as alerts on the app for high and low blood glucose values. Unlike some other CGMs, there is no interference with use of either acetaminophen (paracetamol) or vitamin C. Eversense readings may be used for treatment decisions, but users still must perform fingerstick glucose checks for calibration (although the new system reduces the number).
Findings from the pivotal PROMISE trial with the new second-generation 180-day system were presented June 27 at the virtual American Diabetes Association (ADA) 81st Scientific Sessions by Satish K. Garg, MBBS, MD.
In PROMISE, 180-day sensors incorporating two different chemical reactions were tested, and the overall mean absolute relative difference (MARD) compared with reference values were 9.1% and 8.5%, similar to currently available external CGM devices.
“A Welcome Addition to Real-Time CGM”
“The next-generation Eversense CGM system is safe and accurate for up to 180 days, particularly in hypoglycemia ranges, and allows for a single calibration per day on most days of system wear. With many unique features, this system should be considered a welcomed addition to options available for real-time CGM,” said Garg, of the Barbara Davis Center for Diabetes, University of Colorado, Aurora.
Asked to comment, session moderator Francisco Pasquel, MD, of the division of endocrinology at Emory University, Atlanta, Georgia, told Medscape Medical News: “The fact that it works for a long time, that it’s accurate, and that it can integrate with our smartphones into our ecosystem of technology I think is great news.”
However, Pasquel cautioned that although the infection rate in the study was low, it remains a potential concern with an implanted device. In addition, “some people may not feel comfortable with having something implanted.” But on the other hand, “it may [ease] the burden for some.”
New Chemical in Sensor Further Improves Accuracy
Ninety-six patients wore two sensors, one on each arm. In 53 patients, the second sensor was identical to the primary sensor, and in the other 43 patients, the sensor was modified with sacrificial boronic acid (SBA) to reduce oxidation of the glucose-binding indicator chemistry.
The system prompted users to calibrate with fingersticks twice a day for the first 21 days, followed by once daily.
They were given hyper- and hypoglycemic challenges, and the CGM values were compared with a reference measure (YSI 2300 glucose analyzer).
Of 49,613 paired data points for the primary sensor, overall mean absolute relative difference (MARD) was 9.1%, and 92.9% were within 20% of the reference. The MARD ranged from 7.1% at blood glucose values of 301-350 mg/dL to 9.4% at 40-60 mg/dL. Of 12,034 paired data points for the SBA sensor, the overall MARD was 8.5%.
(For comparison, reported overall MARDs for the external CGM systems Dexcom G6, FreeStyle Libre 2, and Medtronic Guardian Sensor 3 are 9.0%, 9.3%, and 8.7%, respectively.)
On the first day of use, MARDs for the two Eversense sensors were similar, at 11% for the primary sensor and 11.2% for the SBA sensor. However, by day 180, MARD was 10.4% for the primary versus 7.4% for the SBA sensor. Overall, 65% of the primary sensors survived to 180 days, compared with 90% of the SBA sensors.
“Compared with the primary sensor, the SBA sensor performed better. More readings were within 20%…Clearly this biochemical change in the SBA sensor will be the future way to go,” Garg commented.
The SBA sensor also was better at detecting hypoglycemia, identifying 90% of episodes of blood glucose < 60 mg/dL versus 87% for the primary sensor. Event detection at high glucose levels was similar, with both detecting 99% of episodes > 180 mg/dL.
Senseonics has submitted data for both sensor types to US and European regulatory agencies, a company spokeswoman told Medscape Medical News.
A1c didn’t change significantly over the study period, averaging 7.6% at baseline, 7.2% at day 90, and 7.3% at day 180.
“So there was no deterioration in glycemic control and, if anything, a slight improvement,” Garg noted.
There were no serious adverse events related to device insertion or removal procedures, no unanticipated adverse events, and no failure to remove the sensor on the first attempts. There were two mild skin infections, giving infection rates of 1.1% of patients and 0.36% of 558 total insertion and removal procedures.
In June 2018, Senseonics announced a partnership with Beta Bionics to codevelop a “bionic pancreas” system. That project is ongoing, the spokeswoman told Medscape Medical News.
Garg has reported being an advisor for, consultant for, and/or has receive research grants from Medtronic, Novo Nordisk, Zealand, LifeScan Diabetes Institute, Eli Lilly, Lexicon, Dario, Animas, Dexcom, Sanofi, Senseonics, T1D Exchange, National Institute of Diabetes and Digestive and Kidney Diseases, and JDRF. All research grants and honoraria go through his institution. Pasquel has reported receiving research support from Dexcom, Merck, and Insulet, and consulting fees from Boehringer Ingelheim.
ADA 2021. Presented June 27, 2021. Abstracts 149-OR.
Miriam E. Tucker is a freelance journalist based in the Washington, DC, area. She is a regular contributor to Medscape, with other work appearing in The Washington Post, NPR’s Shots blog, and Diabetes Forecast magazine. She is on Twitter: @MiriamETucker.