Anne L. Peters, MD, CDE; Howard C. Zisser, MD


June 26, 2012

This feature requires the newest version of Flash. You can download it here.

New Developments

Anne L. Peters, MD, CDE: Hi. I am Dr. Anne Peters from the University of Southern California. I am here with Dr. Howard Zisser, who is Director of Clinical Research in Diabetes Technology at the Sansum Diabetes Research Institute in Santa Barbara, California. Welcome.

Howard C. Zisser, MD: Thank you very much.

Dr. Peters: Today we are going to talk about something near and dear to my heart or, more importantly, near and dear to the heart of many of my patients: the development of the artificial pancreas. Because you are at the epicenter of this, could you describe it to us and tell us what is happening?

Dr. Zisser: It has been a very active year in the area of the artificial pancreas. We started this project 4-5 years ago, and we have put systems together that we can use in clinical trials. We had to get a lot of regulatory approval to use this in the United States. We have been collaborating with researchers outside the United States. A lot is happening right now. We have done a number of trials,[1,2,3,4,5] mainly inpatient trials in controlled settings, with exercise, meals, and overbolusing and underbolusing. We are trying to automate insulin delivery based on continuous glucose monitoring data. We obtain a set of continuous glucose monitoring data points and we make decisions. Every 5 minutes we get a new data point and make a new decision on what to do with the insulin delivery. So far, the results are very promising. We are trying to eliminate the high blood sugars and the low blood sugars. If we can do that, we can help a lot of people with diabetes.

What Does It Look Like?

Dr. Peters: Help me imagine what you are talking about. What does this artificial pancreas look like? Is it like a box, and how many sensors does it have? How many pieces?

Dr. Zisser: That is a good question because an artificial pancreas isn't any one thing. We all have an idea in our minds of what this would look like. We are not close to the point where we have one widget that we can hand to you, turn it on, and everything will be fine. What we need to do is try to use the technology and the drugs that are available (because they have some limitations) to do the best job that we can. At this point, we don't have the technology or the insulin isn't fast enough to have perfect control all the time. That is why we want to start getting rid of the highs and the lows. Currently we use 2 sensors (it doesn’t matter what brand) and a backup sensor. Some groups are using 2 sensors at the same time and averaging them to try to take some of the noise out of the system. That sensor information goes into another system, such as a laptop computer, telephone, or other platform; it doesn't really matter.

Dr. Peters: These sensors are placed on the back of the arm, or on the abdomen, or where?

Dr. Zisser: They are usually on the abdomen.

Dr. Peters: Okay.

Dr. Zisser: The "brains" of the operation are in whatever device you want to use that makes decisions on how much insulin to give in the next 5 minutes.

Dr. Peters: Such as a cell phone?

Dr. Zisser: It could be a cell phone, or it could be a tablet computer. Sometimes we use a laptop computer because we have to interface, especially in the inpatient trials. It doesn't really matter what platform it is on. When we move to outpatient trials, it will be helpful to have something smaller and start removing receivers and wiring to make it easier for patients to use. That "brain" sends information to the insulin pump. At intervals, every 5 minutes or so, we make note of the decision. We use algorithms that make predictions. The algorithm looks at the data and makes a prediction about where the blood sugar is going and decides what to do from there.

Using the Artificial Pancreas

Dr. Peters: Does it know when you eat?

Dr. Zisser: Another good question. We have done studies and we have a poster[1] at the American Diabetes Association meeting here with a fully automated system. We turn on the system, and the patient doesn't have to interact with the system at all. They exercise, they eat meals, but we don't "announce meals" -- we don't give any premeal bolusing -- and the system works quite well. We also are involved in trials in which we do premeal bolusing, or "announced meals," because the insulin delivery just isn't fast enough. We will be starting a study this summer in Santa Barbara using inhaled insulin. The idea is to give a very small dose of inhaled insulin at the beginning of the meal, and that will act as a first-phase secretion insulin and tell the liver that the patient is going to eat, and then the controller in the subcutaneous system won't have to work as hard. We have some preliminary data out of France with Eric Renard. We are working with him on a DiaPort system (Roche Diabetes Care; Basel, Switzerland) that delivers the insulin in the peritoneal cavity (the abdomen) so the insulin is taken up by the liver almost immediately. Those results look very promising.

Dr. Peters: Doesn't it make more sense to you theoretically to have it more internal than external?

Dr. Zisser: I am a very big proponent of intraperitoneal insulin delivery because it's more physiologic. Insulin is really not designed from a physiologic standpoint to be given under the skin, but the intraperitoneal method takes a little bit more work. You may need minor surgery and more device maintenance, but the devices seem to be very safe. There is not much risk for infection or peritonitis. The patients who have had these systems implanted don't want to go back to the old systems.

Patient Reactions

Dr. Peters: So, they like them?

Dr. Zisser: They like them, and if their blood sugars are tracking low, they can suspend the pump and bounce back much more quickly because they don't have the insulin still being delivered from the subcutaneous space.

Dr. Peters: You would have a much finer-tuned delivery.

Dr. Zisser: Quicker on, quicker off.

Dr. Peters: Your studies are still in inpatients and soon to be outpatients, but when you put somebody on that fully automated system who has had type 1 diabetes for a while, how do they like it? What do they say?

Dr. Zisser: It's very interesting, because we have a select group of patients that we are working with. They are very motivated; they want to know what the latest, greatest technology is, and some of them actually just see it as a vacation day. They come in and we take over all control of their insulin and they don't have to think about it so much. It's interesting, though, to see when we give them meals; they want to do something because they have been doing it for so long. So, they get a sense of freedom, but they probably also feel a little bit anxious at first because they are not in control anymore.

Dr. Peters: Right, I would think that would be a little shocking.

Dr. Zisser: Once they see the results, they think it is great, the best thing that they have ever seen. We will also be doing an outpatient trial at the end of this month. We are going to study 5 patients at the same time using a system developed at the University of Virginia.

Dr. Peters: How long will the trial last?

Dr. Zisser: It's going to be about a 40-hour study.There will be some open-loop sessions towards the beginning. We want some time so we can compare open-loop with closed-loop therapy. We will make some safety checks overnight to make sure that everything is okay. We will be monitoring them remotely the whole time, but we are not going to be right next to them. We may be in the next room or the next building. We will be close by, but the patients are going to be using the system themselves.

When Will It Be Available?

Dr. Peters: Let's say that development goes as it has been going. How long until I can give these systems to my patients with diabetes?

Dr. Zisser: It depends on how you define it. Some people define the "low glucose suspend" as an introductory artificial pancreas. This is a system that is available everywhere except the United States. With this system, when the blood sugar is low for a certain period of time, and the patient doesn't respond to alarms, the pump will be suspended for a couple of hours. We hope that it will be approved in the next year to year-and-a-half. The other thing that we may start seeing before we get a fully artificial pancreas are some remote monitoring capabilities because of how we interact with these devices, and we need to remotely monitor. We may be able to have systems with continuous glucose monitoring devices that family members or parents can monitor and see what is happening to the pump in real time just by logging on to the Internet. We are also developing some glucagon rescue systems for people with hypoglycemia unawareness. These are side projects that we are developing that probably would not have been developed had we not gone down this pathway.

Dr. Peters: You still haven't told me how long from now until you think the artificial pancreas will be ready for primetime.

Dr. Zisser: There are some projections of probable availability in Europe by 2016.

Dr. Peters: Wow.

Dr. Zisser: Maybe in the United States by 2018 for some of the preliminary systems.

Dr. Peters: That would be wonderful. My patients thank you. I love this development. It is obviously going to be too slow for most of us, but I have seen the progression and you have come a long way.

Dr. Zisser: Great, thank you. We are looking forward to it.

Dr. Peters: We are very pleased. This has been Dr. Anne Peters for Medscape. Thank you.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: