Developing an Artificial Pancreas that Delivers GlucagonFriday, August 30, 2013
Software engineer Gil de Paula is developing this device with the capability of delivering a little discussed hormone, glucagon, with the vision of designing a more complete artificial pancreas (AP) system for patients.
By: John Parkinson, Clinical Content Coordinator, DiabetesCare.net
By nature, Gil de Paula is a tinkerer. As a software engineer, his mind is always looking to improve how devices work. In this way, he is always thinking ahead, and this is part of why his company, Pancreum, is developing an AP system that will not only sense glucose and deliver insulin, but will also deliver glucagon.
De Paula says glucagon delivery will mimic a more complete action the body carries out when a person’s blood sugar runs low and glucagon is released to help replenish energy.
His forethought on glucagon combined with his years working on the development of the Medtronic MiniMed Guardian RT CGM and the Insulet OmniPod has given him the experience to undertake such a huge project of designing an artificial pancreas.
It was while de Paula was at Insulet that he met his Pancreum co-founder, Jerry Joseph. Together they have been working towards completion of the first version of their Genesis AP system.
De Paula believes the best and quickest way to get their AP to market is with evolutionary versions of the AP. He says trying to get a fully automated system that senses glucose and delivers insulin and glucagon would take much longer to get FDA approved. For example, De Paula says the first version of his AP might sense glucose but delivery of insulin may take a manual step carried out by the end user.
He explains the evolution of the AP will come from the software and that the hardware will be set up already so adaption to later versions will be easier and cheaper.
As would be expected in such a huge undertaking, there have been challenges along the way. Over the summer they tried to secure crowd sourcing funding from individuals who would be willing to help finance hiring mechanical engineers who could help develop the final components of the Genesis; however, this public funding campaign was unsuccessful and De Paula and Joseph had to come up with an alternative plan, which included finding engineers who were willing to take an equity stake in the company as payment.
DiabetesCare.net sat down to speak with de Paula about the development of the Genesis, the timeline for the project, and funding this major project.
DiabetesCare.net: Why the decision to take this on in developing an artificial pancreas system?
de Paula: While I was at Medtronic and Insulet, I saw that there was a lot of room for technological advancement in individual medical devices. A lot of the bigger companies have a goal to get a product to market and be done with it. As an engineer, I don’t stop thinking about how to make things better. It’s a technology challenge that can be improved vastly. Having acquired the knowledge on the sensor side of things and mini-site pumps, I knew we could begin to work on it.
While there have been lots of challenges, I believe it’s achievable to complete the Genesis AP.
DiabetesCare.net: You have been involved in the engineering and design on a pump and CGM—two important components of what helps to make up an artificial pancreas—how is your Genesis AP going to work?
de Paula: The product is worn on the body and there are sensor devices around it. The modules will do the sensing and the pumping. The hand held controller talks to the device. It can be a custom-made PDA, or depending on regulatory approvals, an app on a smart phone.
All of the components are controlled from the same hand held device. The first version of the product may just be monitoring glucose, and delivering insulin and glucagon individually. We are likely not going to tie it all together in a fully automated artificial pancreas. Otherwise, the FDA may take longer to approve it if we go straight to an automatic system.
The infrastructure for the hardware is going to be there already for the first version. All we are going to get approved by the FDA in the second version and subsequent versions are the advancements to the software.
This should lower the cost of the system too.
DiabetesCare.net: Can you talk about the individual components of the Genesis system?
de Paula: The CoreMD is the brains of the system. It’s home to the microprocessor, Bluetooth component, and where the rechargeable battery is placed. It has a speaker for alerts and alarms. It also has a push button for bolus delivery without the use of the handheld controller—if for some reason the controller is not working.
The GlucoWedge component is a plug-in module, which is disposable and contains the sensor and insertion mechanism.
The Beta and AlphaWedge components are the insulin and glucagon pumps respectively. They are basically identical, but colored coordinated so the person can identify each one.
They contain the reservoir and the insertion and pumping mechanism for the hormone that is being delivered.
DiabetesCare.net: Where are you in development of the individual components and the whole system?
de Paula: We have apps that are ready. For example, the app for the Android smart phone is fully developed; we have a few more features to be added to it, but it’s essentially functional. The IOS app is on its way.
We hope to show the first complete version of the Beta and AlphaWedges there, and also potentially the GlucoWedge this fall at the DiabetesMine summit in Palo Alto, Calif.
Jerry Joesph is a co-founder of Pancreum, a biomedical engineer, and a patent attorney. We met at Insulet, and Jerry has been part of the design team. We have developed everything on our own so far; however, neither one of us is a mechanical engineer, so we needed to hire mechanical engineers to help complete the system.
It`s expensive to hire people so that is why the plug-in modules (BetaWedge and AlphaWedge) have not been fully developed yet.
We tried to raise money to help finish developing them and the interface of the CoreMD.
The main challenges for us are the BetaWedges and AlphaWedges, which contain intricate pumping systems. Once we are done with that, the GlucoWedge is not a huge challenge—once we get the insertion portion complete. The electronics will be done by November.
We just found two mechanical engineers and we were able to convince them to basically work for free in exchange for small equity in the company. We had to go that route as soon as we saw the Indiegogo campaign wasn’t going to work.(Editor’s note: Indiegogo is a crowdfunding platform used to raise money, and Pancreum used it this year to try to raise funds.)
DiabetesCare.net: One of the elements that makes your AP system stand apart from others in development is the ability to deliver glucagon. Why the decision to do so?
de Paula: If you familiar with the research of doctors Steven Russell, MD, and Edward Damiano, PhD, in Boston, you can see they have been doing successive artificial pancreas clinical trials with and without the use of glucagon. Results were not nearly as successful in the non-glucagon trials, and I always thought it didn’t make sense to not have something that ensures your glucose level is at a safe level automatically. Therefore, I thought the Genesis had to have the ability to deliver glucagon. At the time, back in 2008, when I advocated for that technologically, I was speaking like it was science fiction. And even today, there is no FDA-approved glucagon in stable, liquid form.
However, I also knew from my experience that it takes a few years to design a pump and to get it FDA-approved; therefore, I’m sure in that same time period, these companies are going to get approval for glucagon. Why wait for glucagon approval first, which may take a few years, and then tack on a few more years again to design a pump that can be used to deliver the glucagon? Why not design the glucagon pump in parallel to the liquid glucagon?
Somewhere down the road, the two designs will come together and meet. Now, there is glucagon that is in stable, liquid form for five days and they are in the middle of clinical trials. They expect FDA approval by the end of the year or the first quarter of 2014.
We are building a system that is already designed for it, so patients will just plug it in when it’s ready.
Above left: The Vox PDA is a vocal assistance component for the vision impaired and the CoreMD component is shown below that with a USB charging capability; Above right: Screenshot of how it will look using the Android app.
DiabetesCare.net: With the different versions, will people need to replace various components?
de Paula: We decided to design a modular system, which has 3 disposable and one reusable module. This allows the patient to decide what he or she wants to use. It lowers the cost, because the modules are what you dispose, and innovation will come in pumping methods and sensing technology, so all you need to do is come up with a new version of the plug-in module. For example, right now our GlucoWedge is going to be our standard insertion sensor. It senses interstitial glucose levels. It’s an invasive device. If the development of another means of glucose sensing becomes a reality, then all we would need to do is design a sensor like that which fits into our system. That would be the only time to make any hardware changes.
DiabetesCare.net: What are you hoping for in terms of your timeline to complete the Genesis?
de Paula: We hope to have all the components completed by late this fall, but it’s not going to be ready for clinical trial for a while. It is going to be a prototypical rudimentary version of what we want to do.
We are doing everything with our own money and limited resources, so it’s not going to look and feel like the real thing. It will be bigger. I can’t tell you when I go beyond this. I need approximately $2 million to get it to clinical trials and a few more million to get it through clinical trials. And then millions more to apply for FDA approval and make the regulatory requested changes.
It won’t be ready for commercial usage for several more years.
DiabetesCare.net: Would you consider working with other medical device companies to complete the Genesis?
de Paula: Ideally, we would prefer to stay independent so as to give people with diabetes a choice of having another device. However, it’s very hard to get venture capital right now. We have been approached by numerous companies, and one or two of them are very large and they have a lot of capital and have told us they want to help us. However, they want to see us get a little further along in the development process, so there is the possibility that might happen too.
If it comes down to us not being able to get venture capital to stay independent, we’ll need to partner with one of these companies.
To find out more about Pancreum’s Genesis System or to donate to the project, interested readers can email de Paula firstname.lastname@example.org.