Artificial Pancreas Update: Medtronic Checks In
Of all those pursuing a closed-loop artificial pancreas (AP) solution, Medtronic Diabetes already has the most assets required for a successful product. As the market leader in both pump and CGM sales, it is the first and only company to combine a pump and CGM in a single device: the MiniMed Paradigm Revel System. Medtronic also pioneered glucose sensor-to-pump communication, another critical component of a closed loop. Its Paradigm Veo system, available outside of the U.S. for almost three years, includes Low Glucose Suspend, or LGS, technology. LGS first provides two alarms and, if not responded to, automatically shuts off insulin delivery from the pump for two hours when glucose levels go below a limit that is preset by the user. The technology is currently under FDA review as part of Medtronic MiniMed530G system, which was submitted to the agency in June of 2012. With all these assets and experience, Medtronic’s announcement of a closed-loop system must be just around the corner, right?
Not exactly. While Low Glucose Suspend addresses a primary need for a successful closed loop — an automated guarantee that insulin will not be delivered when a person’s glucose is already low — it also showcases Medtronic’s methodical march towards a closed-loop or artificial pancreas developed, tested, produced, and marketed under one roof.
Being the market leader may have a downside when radically different approaches to glucose control are involved. By knowing so much, Medtronic also knows how much can go wrong. All of the other AP systems in development rely on components assembled from different manufacturers. Only Medtronic will bring to market an AP for which it is entirely responsible.
Dr. John Mastrototaro, who leads Medtronic Diabetes’ AP effort as Vice President, Research and Technology, has assembled many pieces of the closed-loop puzzle for 19 years, beginning long before, as he put it, the concept was “in vogue.” He characterized the introduction of Low Glucose Suspend into the company’s technology as one of many “baby steps,” with one successful proof of concept built on another.
Mastrototaros’s first work at MiniMed, before Medtronic bought the company a decade ago, focused on the sensors that supply both pump and patient with information about glucose levels. The sensor’s accuracy is critical, because it provides the raw data that a patient can use to adjust insulin doses in today’s open-loop systems, and which closed-loop systems will use to automate the process once a finished product is on the market. As he sees it, there are two principal challenges to integrating sensor data into a closed loop, one obvious, the other subtle, yet equally important.
First, of course, is the accuracy of the data that is gathered by the sensor every five minutes. Second is an algorithm that can account for and modulate system response to “outlier” high- or low-glucose readings that may be false alarms. This is a complex problem, because one of the issues that users of current CGM technology must address in their current open-loop systems is the human propensity to over-correct, based on glucose readings that temporarily spike or fall due to eating or exercising. It is hard for people to know when to hold off on a bolus; it will be equally hard to express in a formula.
But Mastrototaro said his team is making progress. Assuming the eventual approval of the MiniMed 530G in the U.S., Medtronic has already begun to test, in Europe, a new approach to glucose suspension in an upgrade of that product. It’s based on an algorithm that senses and responds to an impending low before it happens, based on glucose trends identified by the CGM.
. . . there are two principal challenges to integrating sensor data into a closed loop, one obvious, the other subtle but equally important.
This makes data accuracy even more important, and the company also is working on integrating redundant glucose data checks into the next generation of its CGMs. Medtronic’s current CGMs measure the electric current created when glucose reacts with the enzyme glucose oxidase. As a second check, it uses technology developed by the Danish company PreciSense, which it acquired in 2009. The PreciSense sensor measures the amount of glucose binding itself to tiny fluorescent receptors. The receptors give off light that can be measured, quantified, and reported back to the system as levels of glucose concentration. Optically quantifying glucose levels in the tissue will be a second check for the existing sensor, and the two together could be read and correlated by the closed-loop system to ensure that accurate glucose inputs are used by on-board algorithms. While the results of animal and human trials have been encouraging, an approved and marketable dual glucose data check is still some distance from the market.
Naturally, Mastrototaro is reluctant to predict a release date for a Medtronic closed-loop system. But his comments make it clear that Medtronic’s current plan is to test and market a product that combines closed-loop functions overnight with a traditional open-loop system during the day. While nighttime lows leading to severe hypoglycemia are every Type 1’s greatest fear, it’s also true that there is much less flux in glucose levels when people are asleep, since neither exercise nor eating is occurring then. A closed-loop system that keeps glucose levels in a target range overnight would also benefit its user by insuring that prebreakfast readings are as close to normal as possible. This would, in theory, eliminate the “dawn phenomenon” of unexplained elevated glucose, while keeping the user from spending the day trying to correct a morning high, the diabetes equivalent of a dog chasing its tail.
Without question, Medtronic will eventually market a closed-loop system. Ironically, the company that pioneered so much of current pump and CGM technology may not be the first, or even second, company to do so. Perhaps that’s the price of success, but when the eventual Medtronic system appears, it will have been well thought out and thoroughly tested.
Note: Some of the material above was adapted from an article written by Karmel Allison for A Sweet Life, published on April 3, 2012.
Thanks for reading this Insulin Nation article. Want more Type 1 news? Subscribe here.
Have Type 2 diabetes or know someone who does? Try Type 2 Nation, our sister publication.