656-P: Effect of Calibration Frequency on Accuracy of a 180-Day Implantable CGM System

Abstract

Background: The PROMISE Clinical Study was conducted to determine the accuracy and safety of the next generation implantable CGM System. A subset of 43 subjects had a sensor implanted with modified chemistry referred to as sacrificial boronic acid (SBA) , to protect the boronic acid glucose-binding moieties from degradation. This post-hoc analysis investigated the effect of SBA on accuracy through 180 days using different calibration (cal) schemes. Methods: Data from the PROMISE Clinical Study from participants with SBA sensors was run through an algorithm emulator with cals of 1/day, 2/day, or using a dynamic cal scheme switching between 1-2/day, all starting on day 22 (day 1-21 had fixed 2 cals/day) . MARD was analyzed by glucose ranges and days of sensor wear through 180 days. Results: The Table shows the MARD by glucose ranges and cal frequency. MARD (%) by days of sensor wear demonstrated: 9.6, 9.4, 9.2 on days 1-30; 9.5, 8.6, 8.3 on days 31-60; 7.7, 7.4, 7.6 on days 61-90; 8.3, 8.0, 7.9 on days 91-120; 9.6, 8.8, 7.5 on days 121-150; and 7.0, 7.3, 6.9 on days 151-180, respectively, for 1 cal/day, dynamic cal, and 2 cals/day. Conclusions: The SBA sensor maintains an overall MARD less than 9.0% by glucose range and 9.6% by days of sensor wear through 180 days, whether it has a cal scheme of 1, dynamic, or 2 cal per day. By reducing sensor degradation with the SBA chemistry modification, the sensor maintains long-term accuracy with 1 cal/day. Disclosure A.E.Jacquin: Employee; Senseonics, Other Relationship; BrainScope Company, Inc. L.Al-khawi: None. S.Ghosh-dastidar: Employee; Senseonics. K.S.Tweden: Employee; Senseonics. F.R.Kaufman: Consultant; Twin Health, Employee; Senseonics.