Savvy Updates, 4/1/19: Beta Cell Destruction, Patch Pumps (EoPancreas & CellNovo), Integrated Medical Sensors, Insulin Dosing and AI, Hospital-based AP

Researchers May Have Uncovered Stages of Beta Cell Destruction in Type 1 Diabetes, as reported by Trisha Gura of GLU/T1D Exchange, 26 February 2019.

Researchers once believed that the onset of type 1 diabetes led to the complete destruction of insulin-producing beta cells. Recent evidence, however, suggests that most people with type 1 diabetes actually maintain residual beta cells that function long after type 1 diabetes’ onset.

For the study, the researchers studied these individuals with longstanding type 1 diabetes through analysis of samples stored in the T1D Exchange Biobank. Researchers separated participants into three groups – those who were producing high, medium, or low amounts of C-peptide. Surprisingly, researchers found that nearly all study participants were producing proinsulin, even those with low C-peptide, which should correspond to levels of its parent proinsulin. Also, only those who could convert proinsulin into insulin, as measured by C-peptide levels, were observed to increase proinsulin production in response to a meal.

From these results, the researchers speculate that there may be an observable hierarchy of beta cell dysfunction. At the onset of type 1 diabetes, beta cells can still process proinsulin into C-peptide and insulin. Over time, the cells may lose the ability, causing them to be targeted for destruction. As beta cell destruction progresses and C-peptide levels decrease, the ability to increase proinsulin after meals is reduced.

EoPancreas: A New Patch Pump Closed Loop System in the Works was reported by Mike Hoskins of DiabetesMine, 20 March 2019. Indeed, exciting news: the FDA has granted what it dubs “breakthrough device designation” to a future closed loop system that combines a continuous glucose monitor with an insulin patch pump to rival the tubeless OmniPod.

On March 14, the FDA gave South Korean company EoFlow this fast-track designation for its Automated Insulin Delivery (AID) system called EoPancreas. This is the fourth time FDA has granted the new breakthrough device designation to a diabetes closed loop system since it was created in 2016 — Bigfoot Biomedical received it in November 2017 for their future AID tech, and Medtronic received it with pediatric approval of the Minimed 670G last summer and then again in February 2019 for a future “personalized closed loop system” the company is developing. While EoFlow isn’t as well known as some others in this space, it’s a company we have seen over the years displaying its patch pump and connected mobile app at conferences like the ADA’s annual Scientific Sessions. The tubeless pump itself is called the EoPatch, and rumor had it they were working on a CGM but there hasn’t been much detail available to date.

EoFlow is a South Korean company founded in 2011 that now has an office in San Jose, CA. Its founder and CEO Jesse Kim ran a semi-conductor startup in Silicon Valley after graduating from MIT and eventually started up EoFlow as a university incubation center in Seoul. Since October 2017, EoFlow’s President Luis Malave (a former Insulet executive over a decade ago and Minimed’s R&D director during the ’90s) has been part of the team working out of the California office.

Referred to as the EoPatch, it’s a rectangular pager-sized insulin pump that adheres to the skin without the need for plastic tubing like traditional pumps use. It’s fully disposable and waterproof, lasting 72 hours on the body and holding up to 200 units of insulin. The design that will ultimately come to market may vary to some degree, but the form factor remains mostly the same. The existing first-generation EoFlow patch pump is quite a bit thinner (at 9.9mm x 32.4mm x 12.9mm) than the OmniPod (39mm x 52mm x 14.5mm). The EoPatch also uses a 30-gauge stainless steel needle inside to help minimize occlusions (clogging). But because the CGM would be built into the patch pump itself, chances are it would add a bit of thickness, roughly half a centimeter, we’re told. Malave also tells us they expect to launch just the patch pump in the U.S. first, as a sort of first-generation before adding in smart algorithms and eventually the built in CGM technology.

This article also touches on these patch pumps in development:

DiabeLoop (a French startup) with its DGLB1 hybrid closed-loop system … I wrote about this as the multi-colored Kaleido hybrid patch-tubed pump a few months ago;
Omnipod Horizon, a patch pump version of a closed loop system
Roche Solo (FINALLY): in 2018 Roche launched the Solo micro-pump in Europe. It’s a thin little semi-disposable device that is just a tiny bit larger than the OmniPod, but notably it’s detachable (!) and you can bolus directly from the little Solo patch pod itself using built-in buttons. Roche is also working to integrate CGM technology such as the implantable Eversense CGM, and will eventually make that part of its own closed loop system in development.
Lilly Diabetes is also working on a hybrid patch-tubed pump that “sort of resembles a round tobacco tin that fits in the palm of your hand.”
Tandem’s t:sport MiniPump that is a disposable, stick-on-your-body device that requires a short connector.
SFC Fluidics is also developing an open-protocol patch pump closed loop system.

Cellnovo Stops Manufacturing and Commercial Operations according to BusinessWire.com, 28 March 2019.

Cellnovo Group (“Cellnovo” or the “Company” CLNV:EN Paris) (Paris:CLNV), a medical technology company marketing the first connected diabetes management platform composed of an insulin micro-pump communicating via Bluetooth® with a locked-down Android^TM smartphone, today announces that the group will discontinue all manufacturing and commercial activities, effective immediately.

The administrators appointed at Cellnovo Ltd., the group’s UK subsidiary, have concluded an initial analysis and have actioned a restructuring plan to focus all activities on identifying strategic partners and financial investors. As a result, a core group of existing employees from the Research and Development department will be tasked with pursuing these objectives. Production and commercial operations, including sales and marketing activities, have been discontinued.

As of this week, systems will no longer be made available to new patients. The commercial team will be working closely with partners and distributors to ensure as smooth a transition as possible for patients and their healthcare teams.

Sophie Baratte, Chief Executive Officer of Cellnovo, commented: “This is not a decision that was taken lightly but it is the only option that is available to us today. Our biggest priority has always been patients’ well-being and we will continue to do our utmost to support a smooth transition for patients and clinicians. We are grateful to all those who helped us to bring our innovative diabetes management platform to the market, particularly the clinicians and patients who believed in the benefits of this novel system. We are also thankful to our distributors and partners who shared our passion and brought their professionalism to the service of our commercial activities. Unfortunately, we were unable to withstand the pressures of the competitive environment and the challenges of rolling-out a breakthrough system at a sustainable cost. We will now focus our efforts on leveraging the potential of a unique product and the value created over many years by an exceptional team.”

Read the release: Cellnovo Stops Manufacturing and Commercial Operations

Watch Integrated Medical Sensors CEO Muhammad Mujeeb-U-Rahman describe their small, user-insertable, and low-cost wireless glucose sensor.

IMS (a multidisciplinary team, consisting of Caltech graduates with domain expertise in microelectronics, nanotechnology and electrochemical sensors, industry leader in implantable glucose sensor technology, leading endocrinologists, serial entrepreneur business leader with experience in multinational corporations, and industry experts in embedded systems and mechanical design) has developed world’s first fully-integrated, extremely miniature (less than half of a sesame seed) wireless electrochemical glucose sensor with 10x longer projected lifetime at 10x lower projected cost compared to the available solutions.

The revolutionary IMS sensor is implanted few millimeter under the skin to measure glucose from the interstitial fluid using an integrated electrochemical sensor and relays the real-time data wirelessly to the patient and a secure database (accessible by caregivers, doctors) via a wearable wireless transmitter and a smart reader.

And read more: Integrated Medical Sensors

Better Insulin Dosing via Machine Learning and Bolus Dosing Skills that you Didn’t Learn in the Doctor’s Office are VERY interesting articles by Martin Hensel and Nicole Rubenstein for InsulinNation.com, 26 and 28 March 2019.

Bolus Dosing Skills that you Didn’t Learn in the Doctor’s Office account for the following issues:

Issue #1 – Most meals are not JUST carbohydrates; they are a mixture of fats, proteins, and carbs. These three nutrients digest at different rates and because of this, your insulin may not match the curve of the glucose rise from your meal.
Issue #2 – Free fatty acids cause insulin resistance. Because of this, insulin requirements would need to be higher to handle the glucose load from a meal that contains carbs but is also high in fat.
Issue #3 – Glycemic index (GI) is not synonymous with grams of carbohydrate for food. Therefore, some meals will produce a higher more rapid blood sugar spike (high GI) while other meals cause a more gradual rise in glucose (low GI), even though the meals may contain the same amount of carbohydrate.
Issue #4 – Individuals are all over the map with their carb counting skills. However, it’s interesting to note that precision is likely more impactful than accuracy when it comes to carb counting and improving A1C. Precision is valuable because consistency, whether wrong or right, enables a more reliable insulin-carb ratio.

Better Insulin Dosing via Machine Learning, and treating T1D is complex and Quin is collecting the data to do it better.

Each person learns over time and seeks to optimize their insulin dosing based on a number of factors, including not only foods they have eaten, insulin sensitivity and insulin-on-board but they often consider factors such as their recent sleep and exercise, stress levels, sickness, menstruation, and altitude.

Cyndi Williams, Co-Founder and CEO of Quin — https://quintech.io — discussed their Insulin Intervention Taxonomy and how these data sets produce better T1D insulin dosing recommendations. Her core mission is to systematically formalize and classify this decision data to enable better future decisions, and stratify people according to their decisions and outcomes for more personalized medicine. To that end, Quin is building an insulin management app that helps people make quick decisions to manage food and activity by tracking what may be affecting them, and recalling what’s worked for them in the past.

Read more:

I know this is a LOT of reading … but just one more, because it is interesting! One of our fears as T1s is being hospitalized, where we lose the ability to control our own diabetes management.

Admetsys Hospital-Based Artificial Pancreas is Making Progress Towards Market was reported by Sony Salzman on T1D Exchange / GLU, 28 January 2019.

The critical care unit is one of the most high-tech places in any hospital, outfitted with state-of-the-art medical equipment to quickly handle most medical issues. Despite this, nurses still manually analyze blood samples and adjust insulin infusion rates of patients. This multi-step process, which can consume a large amount of time of the nursing staff, could in theory be performed perfectly by a machine.

This challenge inspired father-and-son duo Tim and Jeff Valk to launch Admetsys, a company focused on creating an artificial pancreas glucose control system designed specifically for hospital and surgical care. The Admetsys system is a dual-hormone automated glucose control system that would automatically adjust insulin and glucagon delivery based on glucose levels.

The duo’s journey started over a decade ago when Tim Valk, a practicing endocrinologist in Orlando, Fla, received a 3 a.m. phone call from one of his nurses. She lamented that too much staff time was being squandered on the tedious task of patient blood sugar management.

“Why don’t you fix this?” the nurse demanded, before abruptly hanging up.

Instead of feeling chastised, Dr. Valk was inspired. He called Jeff, an engineer and self-described computer “geek”, and asked if an artificial pancreas at the bedside would even be possible; Jeff believed it was. They teamed up with Glenn Robertelli, a medical device expert who previously worked at Johnson & Johnson, and the trio soon launched Admetsys. A decade later, their quest to bring this system to market is making progress.

My first reaction was: “Oh wow, she thinks blood sugar management is tedious for a short period of time … try it every moment of every day!” BUT … she was calling attention to a serious problem, and THANK GOODNESS, Dr. Valk saw the need.