Seattle-based Know Labs conducted a study evaluating its proprietary radiofrequency (RF) dielectric sensor. The sensor non-invasively measured blood glucose in participants with prediabetes and type 2 diabetes. It used venous blood as a comparative reference.
Know Labs reported a mean absolute relative difference (MARD) — a measure of accuracy — of 11.1%. Comparatively, market-leading minimally invasive CGMs from Dexcom and Abbott have demonstrated higher levels of accuracy. Dexcom’s G7 previously demonstrated a MARD of 8.2%, while Abbott’s FreeStyle Libre 3 came in at 7.6%.
However, Know Labs’ non-invasive offering remains an interesting prospect, given there are no FDA-cleared CGMs of this kind at present. Larry Ellingson, a Know Labs board member, highlighted the importance of this initial data.
“Achieving this level of accuracy in the first study using a blood reference device is truly remarkable,” Ellingson said in a news release. He previously served as chair of the board of the American Diabetes Association.
The company last week raised $12 million to work toward bringing its non-invasive CGM technology to market. The company unveiled KnowU, its wearable CGM set for FDA submission, last month. KnowU’s sensor uses spectroscopy to direct electromagnetic energy through a substance or material. Through this, it can capture a unique molecular signature. The technology integrates into wearable, mobile or bench-top form factors.
Dr. Virend K. Somers of the Mayo Clinic served as author and co-investigator of the clinical research. Somers presented the data at the Advanced Technologies & Treatments for Diabetes (ATTD) conference in Florence, Italy.
More about the study results
This study marks the company’s first clinical research protocol involving people with diabetes and using venous blood as a comparative reference.
Know Labs’ RF sensor continuously scanned forearms during 21 three-hour sessions involving a 75g oral glucose tolerance test. Investigators collected venous blood samples using a peripheral intravenous catheter every five minutes. They analyzed blood using an FDA-cleared blood glucose hospital meter as a reference device.
By including participants with type 2 diabetes, the study significantly increased the number of data points outside the normoglycemic range, specifically within the hyperglycemic range. The sensor demonstrated similar accuracy levels across both ranges.
The company said the results suggest that the sensor, paired with machine learning algorithms, holds promise for non-invasive measurement.
Know Labs presented the results as part of a larger, now-completed clinical trial of up to 100 participants.
“Despite significant efforts in the development of non-invasive blood glucose monitoring solutions, delivering a highly accurate, economical, FDA-cleared non-invasive continuous glucose monitor (CGM) still remains to be seen,” Ellingson said. “These results from ongoing clinical research reflect the impressive and continued stability of Know Labs’ novel sensor technology and algorithms, bringing the company one step closer to delivering its FDA-cleared non-invasive CGM to the masses.”
