A 180-day implantable continuous glucose monitor (CGM) was found to be both safe and effective, according to the prospective PROMISE study. In an analysis of 181 adults with diabetes, Senseonics' next-generation Eversense CGM system implanted in the arm for 180 days demonstrated accuracy, with an overall mean absolute relative difference (MARD) of 9.1% between clinic reference values and the device's primary sensor, based on nearly 50,000 paired CGM points, reported Satish K. Garg, MD, of the University of Colorado in Aurora. When measuring all glucose ranges, this primary sensor had a median absolute relative difference of just 6.7% and a reading within 20/20% of reference values of 92.9%, Garg explained at the virtual American Diabetes Association (ADA) Scientific Sessions. The primary sensor was most accurate when measuring euglycemic ranges, with the highest MARD reported with hypoglycemic ranges: 40-60 mg/dL: 9.4% 61-80 mg/dL: 8.8% 81-180 mg/dL: 9.0% 181-300 mg/dL: 7.7% 301-350 mg/dL: 7.1% 351-400 mg/dL: 8.0% The implantable Eversense CGM system is currently FDA approved for 90-day use, and is already approved for 180 days of wear in Europe. When testing CGMs, smaller MARDs indicate more accurate sensors. Generally, anything less than 10% is considered very good. A previous study, presented at a past ADA annual meeting, demonstrated accuracy of this 180-day system in a young population with type 1 diabetes, yielding an overall MARD of 9.4%. The PROMISE study, which was conducted across eight clinical research sites, specifically tested two different sensors used with the CGM system. Of the 181 participants, all had the primary sensor implanted into one arm; one half had an identical sensor placed into the other arm, while the other half had a different sensor placed into the second arm -- a Sacrificial Boronic Acid (SBA) sensor -- which has a specific chemical makeup aimed at reducing oxidation of the glucose-binding indicator chemistry. Participants were seen at 10 clinic visits that took place during the 180 days of use, each visit lasting up to 10 hours. At these visits, participants underwent hyperglycemic and hypoglycemic challenges to test the full 40-400 mg/dL blood glucose range. CGM readings were compared with a reference glucose reading from a Yellow Springs Instruments (YSI) 2300 glucose analyzer. The cohort was generally split between men and women, with an average age of 49. More than 90% of the cohort were white. The average duration of diabetes was 22 years. Average baseline HbA1c was 7.6%, and 70% had type 1 diabetes. During the first 21 days of CGM use, the device prompts for two calibrations per day, which later drops to one calibration per day after day 21. Average HbA1c improved across the cohort, dropping from 7.6% to 7.3% at day 180. Looking specifically at how each sensor performed, only 65% of the primary sensors survived up to the full 180 days of implantation. Nearly all survived to day 90 (98%), 90% were still working by day 120, and only 74% survived to day 150. The SBA sensor performed far better, as 90% lasted the entire 180-day wear -- the most important difference between the sensors, according to Garg. When broken down, 96% of SBA sensors survived to day 90 and day 120, and 94% survived to day 150. SBA sensors also had a lower overall MARD -- 8.5% -- based on 12,034 paired points. This sensor seemed to stabilize with use, starting off with a day 1 MARD of 11.2%, but ending with a MARD of 7.4% at day 180. Also, about 94% of CGM readings were within the 20/20% of YSI values. They also were slightly more accurate in regard to hypoglycemia, with a 7.5% and 7.7% mean absolute difference at ranges of 40-60 mg/dL and 61-80 mg/dL, respectively. The SBA sensors also performed better in alerting the patient to hypoglycemia (glucose of 60 mg/dL), with a true alert rate of 73% versus 68% with the primary sensor. "Clearly, this biochemical change that has been made in the SBA sensor is probably the future way to go," Garg noted. "The majority of the patients can have 6 months of data available to them with one particular implant." Overall, there were no serious adverse events reported with insertion or removal of the devices, and there were no unanticipated adverse events during the total 558 insertion and removal procedures. There were also no failures to remove the sensor on the first attempt. Two mild skin infections occurred, affecting only 1.1% of participants and only 0.36% of the total number of procedures. "[The next-generation Eversense CGM system] is safe and accurate for up to 180 days, especially when you talk about the SBA sensors, particularly in the hypoglycemia ranges," Garg highlighted. "It's also important to note that this particular sensor has no issues with Tylenol or paracetamol or with vitamin C," he added. "As many of you know, the Medtronic system still has issues with paracetamol or Tylenol, and [FreeStyle] Libre has issues with vitamin C. It's only the Dexcom sensor that doesn't have issues ... but definitely the Eversense system doesn't have any interference." Source
