This was not an easy year for the small medical device company Echo Therapeutic (ECTE). In June, its stock went through a 1:10 reverse split and more recently its management team was fired. The interim CEO Robert Doman has been working hard to turn the company around.
As part of this process, the company announced on November 26th a positive CE Mark regulatory trial result for its Symphony CGM (continuous glucose monitoring) system. The company has shown that CGM has an error rate of 12.5% by the Mean Absolute Relative Difference (MARD) method and a 97.9% clinically accurate glucose detection by the Continuous Glucose-Error Grid Analysis (CE-EGA). Investors cheered the news and bid the stock up on the day of the press release by 72% to $4.65.
Despite the positive data neither the press release nor conference call highlighted the true implications of the data. The conference call basically repeated the information in the press release and the lack of any audience questions may be more of a lack of interest from analysts than belief in the data. After comparing ECTE results with previous studies, I found a few issues that may prevent the Symphony CGM system from being adopted by European ICU doctors.
Drug interference could be a serious issue.
The trial recruited 32 subjects but 3 of the patients were excluded from the analysis because an IV formulation of acetaminophen interfered with the sensor of the Symphony CGM system. The company has not stated if other drugs could cause a similar issue but this should not be overlooked. Acetaminophen is widely used in hospital ICUs and this led to the exclusion of almost 10% of the patients. If these patients were included in the analysis, the error rate would be much higher and accuracy questionable. As we know, the most important data come from the ITT (intent-to-treat) population and not selected data from a mITT (modified intent-to-treat) cohort.
Safety data were missing.
The Symphony CGM system is a transdermal detector that requires a skin treatment to remove the outer layer before it sticks to the skin for the duration of the monitoring session. The press release stated that "prepared skin sites were inspected for redness or other undesirable effects immediately following sensor removal, and again 7 days after sensor removal", but management decided not to release any information from this analysis. While there could be nothing to show, it is still an unknown at this point.
Previous studies showed that the system could not accurately detect abnormal glucose levels.
A doctor is most concerned about abnormal glucose levels, where either hyperglycemia or hypoglycemia is dangerous. The press release, however, failed to disclose the accuracy of the test for these vital categories (hypoglycemia, euglycemia, and hyperglycemia). In addition, we were not informed about the glucose levels of the 29 subjects included in the analysis.
A recent company presentation dated November 8th did, however, show that the Symphony CGM system could not accurately detect abnormal glucose levels, particularly hypoglycemia. The following tables are copied directly from the company presentation.
a) MR I Study
b) MR II study
c) MR III study
d) MR IV study
(click to enlarge)
The tables clearly seem to show that the system has lower accuracy when the glucose level is either too high (hyperglycemia) or too low (hypoglycemia). In particular, the accurate detection rate for hypoglycemia is as low as 33.9%, which means an error rate as high as 66.7%.
Symphony CGM system may be too cumbersome to use
The supposed appeal of the CGM system is the time it saves the nurses who have to monitor the glucose levels of ICU patients. Currently it takes about 5 minutes for a nurse to measure the blood glucose concentration using a standard glucose analyzer. The Symphony CGM system requires careful skin treatment using the accompanying Prelude SkinPrep to increase the permeability of the skin that allows for the extraction of the analyte. Unlike a blood sample that is structurally homogeneous, each person has a different skin structure and permeability. The symphony CGM system needs careful calibration before each and every detection. In an ICU where blood samples are readily available (can be taken from arterial line catheters), the cumbersomeness of the Symphony system could be a deterrent to nurses rather than a boon. That being said, I am not sure about usage issues, so take it for what it is.
The US FDA requires rigorous clinical trials to demonstrate safety and efficacy before a device can apply for PMA (pre-market approval). In contrast, the European CE Mark, which uses a different process and requirements, does not. In general, the CE Mark is a much easier path than PMA. This explains why small device companies such as Cytosorbents Corporation (OTCQB:CTSO), Mela Sciences (MELA), and Cytori Therapeutics (CYTX) have received CE Mark approvals long before they were, or will be able to, file applications for PMA. As such, a CE Mark does not mean that doctors or hospitals will use the device. In this regard, CTSO, MELA and CYTX all serve as good examples.
In summary, the data in last week's press release are similar to the previous studies with some key efficacy and safety data missing. Given the dubious efficacy and difficulty of use, I doubt the Symphony CGM system will be used in EU hospitals even if/when Echo Therapeutics receives a CE mark.