FDA panel to make sense of implanted glucose meter

Senseonics best-in-class claims to be tested when blood glucose monitor goes before regulator.

The Eversense blood glucose meter will come before a US FDA committee at the end of this month, and with its developer, Senseonics, claiming that pivotal trial data make it best in class in terms of accuracy an approval recommendation is a strong possibility.

Unlike the other meters on the market, though, Eversense is implanted. Patients must undergo a minor incision-and-insertion procedure four times a year, and must also take dexamethasone to guard against an immune response at the sensor insertion site. These drawbacks will have to be weighed against the efficacy data and, given that Eversense is Senseonics’ only product, the company needs a yes vote.

Eversense is implanted in the upper arm by a doctor. The patient receives a numbing injection and the device is inserted under the skin through a small incision, using a special tool. Like all continuous glucose monitors (CGMs), it detects glucose in the interstitial fluid and extrapolates blood sugar from there – but it works in a different way to current devices, using a fluorescence-based measurement technique where all the others rely on electrochemistry.

Roche picked up distribution rights in several European countries, but when it comes to the US Senseonics is on its own (Senseonics hooks Roche for implantable glucose monitor, May 26, 2016).

Eversense's US approval trial, Precise II, showed that the device had a mean absolute relative difference (Mard) of 8.8% over the full glycaemic range for the entire three-month sensor duration. Mard is the difference between the CGM reading and a reference lab test done on a sample of the patient’s blood – the smaller the difference, the more accurate the sensor.

The 8.8% difference is indeed better than the Mards seen with other CGMs, according to published data. All the devices in the table below are already approved in the US with the exception of Nemaura’s SugarBeat, which reported pivotal data last month.

CGM accuracy comparison
Company Device Mard (%)
Senseonics Eversense 8.8
Dexcom G5 9.0
Abbott Laboratories FreeStyle Libre 11.4
Abbott Laboratories FreeStyle Navigator II 12.3
Dexcom G4 Platinum 12.5
Medtronic EnLite 13.6
Nemaura Medical SugarBeat  13.8
Sources: company websites, published data.

Precise II was a small trial, in just 90 patients, and included no control group. But Eversense’s safety profile was pretty clean, with 14 device or procedure-related adverse events in seven subjects, the most frequent being mild bruising – hardly surprising given the insertion procedure.

Sense and sensitivity

There is one minus common to CGMs that Eversense has and one of its competitors does not: calibration. Patients must perform fingerstick blood tests twice a day to maintain Eversense’s accuracy – the same as Dexcom’s G4 and G5, and rather better than Medtronic’s EnLite, which must be calibrated three or four times a day.

But Abbott’s FreeStyle Libre, launched in the US in September, allows patients to go a day and a half, on average, without having to jab a needle in their finger (Dexcom crashes as Abbott FreeStyles its way to the US, September 28, 2017).

Perhaps this is more a commercial concern for Senseonics than a question that will exercise the advisory committee on March 29.

In the end the panel will have to make the same calculation the patients themselves will make if the device is approved: is the high accuracy and longer wear time – three months, compared with a maximum of 11 days for rival sensors – worth undergoing minor surgery four times a year and having to take immunosuppressants?

Share This Article