Advertisement
"This study represents another major advancement in our Prelude andSymphony development programs," stated Patrick Mooney, M.D., Echo's Chairmanand CEO. "We have now successfully used Prelude, our new needle-free skinpreparation system, in combination with our non-invasive Symphony tCGMtechnology. Data from this study demonstrate that Symphony, using Prelude, issafe, accurate and reliable at monitoring glucose levels. We look forward tocontinuing progress with our Symphony development program throughout theyear."
Advertisement
Study Design
After Prelude skin permeation, Symphony tCGM biosensors were applied toten subjects with Type 1 or Type 2 diabetes. Venous reference blood sampleswere taken from intravenous lines at 15-minute intervals for 24 hours andmeasured on a YSI 2300STAT PLUS laboratory analyzer. At the conclusion of the24-hour observation period, the test skin sites were inspected for redness orany other undesirable effects.
Analytical Methods
The primary statistical analysis tools used to evaluate the performance ofthe Symphony tCGM System relative to the reference measurements were theClarke error grid, mean absolute relative difference (MARD) and linearregression, each of which are widely used to evaluate the performance ofglucometers. The Clarke error grid is a plot of all data pairs categorizedinto five discrete areas: A, B, C, D and E. The A and B areas are the mostclinically desirable zones and D and E are the least clinically desirablezones. Devices with a higher combined A and B zone percentage (closer to 100%)and lower combined D and E zone percentage (closer to 0%) are considered tohave better performance. Continuous Glucose Monitoring (CGM) systemperformance, including tCGM system performance, is competitive if at leastninety-five percent (95%) of the data points fall within the combined A/Bzones, along with negligible or no data points in the combined D/E zones.Mean absolute relative difference (MARD) is a standard error calculation toolthat is used to measure the average absolute value of the relative (orpercentage) difference between two measurements. A low MARD error, below 20%,is generally considered competitive.
Study Results
Using 1,292 reference blood glucose measurements from the ten subjects inthe study, Clarke error grid analysis of the study data showed that Echo'sSymphony tCGM System had approximately 99% of the data in the combined A/Bzones, with 76.4% in the A zone and 22.4% in the B zone, and only 0.2% and0.9% in the C zone and D zone, respectively. The MARD for the study was13.8%. The correlation coefficient (R) for the data was 0.89. There were noadverse events reported from Prelude skin permeation or the Symphony tCGMbiosensor.
Conference Call and Webcast Information
Echo will host a conference call today at 11:00 AM ET to discuss the datafrom its recent clinical study and to provide an update on its Symphonydevelopment program. To access the conference call, please dial 866-394-4716(domestic) or 706-634-5164 (international) and reference the conference ID58102479. A replay of the call will be available from 12:00 PM. ET on July 29,until August 12, 2008, at midnight. To access the replay, please dial
