What if we treated high glucose not by giving a drug, but by removing glucose itself?
A study in Science this week describes ECHO, an implantable device designed to do exactly that, without insulin and without medication.
How it works:
The implant sits under the skin and contains an electrocatalyst made of platinum decorated gold nanoclusters embedded in porous graphite.
When wirelessly powered at low power (0.75 volts), the catalyst selectively converts blood glucose into gluconate. Gluconate carries less energy and is cleared by the kidneys into the urine, so both glucose and calories leave the body.
Because the reaction is driven electrically, the rate of glucose removal is adjustable. Turn the current up, remove glucose faster. Turn it off, removal stops.
The closed loop comes from pairing it with a standard continuous glucose monitor. The CGM detects glucose above a set threshold, the implant activates, glucose falls, and the system switches off. In this study it was also smartphone controlled.
What happened in mice with type 2 diabetes and obesity:
Glucose returned to the normal range within 1 hour of activation. Repeated treatment maintained control for 6 weeks with no detectable hypoglycemia. As glucose was removed, metabolism shifted toward fat use. Body weight and triglycerides fell, while free fatty acids and ketone bodies rose. No adverse physiological, inflammatory, or behavioral effects were detected over those 6 weeks.
This is early stage, mice only, 6 weeks of data. Human questions are unanswered: catalyst durability over years, tissue reaction to a permanent implant, kidney handling of a sustained gluconate load, and whether the glucose removal rate scales from a mouse to a human.
We already have effective human therapies with outcome data, from metformin and SGLT2 inhibitors to GLP-1 receptor agonists and automated insulin delivery. ECHO is not a replacement for any of them today.
It is a proof of concept for a different category: closed loop diabetes care that does not deliver a drug at all, but removes the substrate itself.
science.org/doi/10.1126/scie…