Aaron Kowalski, PhD, assistant vice president of treatment therapies at JDRF, walks us through where we are with the promising technology including what challenges remain and a timeline for when we might see a closed loop device in the United States.
By: John Parkinson, Clinical Content Coordinator, DiabetesCare.net
For Type 1 diabetes veterans, they have been through a roller coaster of emotions regarding a possible cure for the disease. For years, they were told a cure was coming, only to remain disappointed when nothing surfaced. And in some ways, this has affected T1s’ outlooks in how they view news of potential cures or treatments.
Now comes word of the artificial pancreas and the hope this promising technology holds. For some type 1s, this “hope and promising” talk might drudge up some old feelings, and when you talk to some T1s, they take a pragmatic stance to the question of whether they are excited by the artificial pancreas. They speak of how the technology is exciting, but they also have some of that “I’ll believe it when I see it” attitude sprinkled in there.
Certainly, no one is equating the artificial pancreas to a cure, but JDRF’s Dr. Kowalski, who is the lead scientist on the organization’s Artificial Pancreas Project, and a T1 person himself, says this technology is truly a game changer.
Dr. Kowalski says this technology will greatly affect peoples’ quality of life, and will be able to do so primarily in two ways: daily self management--keeping people within an accepted glucose range longer so that in the long term they can delay or avoid the possibility of complications altogether; and more acutely with safety, keeping people from experiencing severe hypoglycemic incidents when sleeping through the night.
JDRF is the largest charitable funder and advocate of diabetes science worldwide, helping to set the global agenda for diabetes research. Therefore, with this type of international reach, JDRF knows of, or is involved in, all active artificial pancreas research going on throughout the world and speaks from a position of authority.
JDRF has been working with a wide range of professionals from different industries including researchers, clinicians, engineers, and mathematicians on the artificial pancreas project. Dr. Kowalski says there are products being used internationally that are more advanced than the technology used in the U.S., and JDRF is working with the U.S. Food and Drug Administration (FDA) to alleviate any potential roadblocks.
While Dr. Kowalski understands firsthand the history of broken promises of finding a cure, and experiences present day frustrations in dealing with governmental inertia, he remains optimistic about the artificial pancreas. He sees the technology as an evolutionary process taking steps along the way to become more sophisticated with an eventual endgame product that will have the ability to provide multi-hormonal delivery directly to the body.
In this part one segment of a two part extensive interview with Dr. Kowalski, we discuss JDRF’s expectations for the artificial pancreas, one of the existing pumps today that is on the market internationally and is able to perform what is considered the first step in the artificial pancreas project, and insights into ongoing dialog JDRF is having with the FDA in trying to bring this device and future ones to the U.S. market.
DiabetesCare.net: Can you define what the artificial pancreas means to JDRF in terms of what the technology will be able to do?
Dr. Kowalski: Many people have a different idea about what an artificial pancreas is to them, so I’ll give you our feelings and how we are working on this strategically. To us, we obviously want to replace what the body has lost due to type 1 diabetes, including insulin and glucagon. In the future, the goal is a mechanical means to replace those hormones and restore balance.
What we have tried to do is create our “closing the loop” roadmap. This includes steps where we think some mechanical approaches could be developed that would be meaningful both clinically and to patients in the near term while we drive towards a very sophisticated hormone-driven artificial pancreas.
This is happening in parallel, but in different time frames. For example, in the near term, in JDRF-funded studies, we have seen you can reduce hypoglycemia by about 75% using a predictive algorithm. That is the first step. This is simply taking glucose in real time and turning off an insulin pump, which could reduce a lot of hypoglycemia. You can turn it off once you get to a certain number or you could be even smarter about it and reduce doses as a person is going down.
I would argue one of these products already exists. Specifically, Medtronic has a product called the Minimed Paradigm Veo insulin pump. It is available all over the world but not available in the United States right now.
Simply shutting off the insulin when glucose is low has really been done. If you have someone who is severely low we don’t want to pump more insulin into their bodies and that’s what happens with today’s pumps. This product deals with this simple problem of hypoglycemia and likely can prevent people from dying.
Ultimately, what we are trying to do is take today’s pumps and sensors and maximize their benefits using computer algorithms. In the near term, I’m convinced we can do that.
In the medium term, it would start to dose insulin. Here is where it gets a little nuanced, but we have a number of approaches that we are trying to use.
DiabetesCare.net: You had mentioned the Minimed Paradigm Veo insulin pump. Understanding that it is being used outside the United States, what has been the holdup for its use in the U.S.?
Dr. Kowalski: We are working with all the manufacturers, and we are counting on them to deliver these devices, so we try to be as agnostic as possible.
However, we are deeply involved in this issue, and in our discussions with FDA, it has been challenging. We feel patients in the U.S. should have access to the product. In draft guidance, FDA has been asking for studies which JDRF and many other diabetes organizations have deemed unreasonable. JDRF has formally submitted comments stating what we believe is a more reasonable pathway.
Overall, we fund a number of academic sites, which partner with world-class doctors, mathematicians, engineers, and scientists to do artificial pancreas-based research. Many of them are working on the various permutations of the artificial pancreas including hypoglycemia systems, dosing systems, and multi-hormone systems. We have a number of them here in the U.S., and also in Europe, Israel, and Australia.
What we have seen in the hypoglycemia research, which is turning down insulin, we believe the FDA is focusing on the wrong issue. FDA has put out a draft guidance document, which gives guidance to industry and the medical community on what they think is necessary to get approved. In this guidance, they stated we needed to do a series of studies that everybody in the community believes is excessive. What it gets at is what you need to do to approve a system that turns off insulin if someone is low. The guidance says you need to demonstrate people get low less frequently. This sounds fairly reasonable. However, if you look at this a little more carefully, with a low-glucose suspend product the pump turns off when the person generally already is low, so doing this type of study is very difficult in trying to show you are reducing hypoglycemia.
Then there are some additional challenges if you get into the study design aspects and what they have asked for. What JDRF has said is, ‘why hold this product up? What would the risk be to doing this?’ A hypothetical risk is if you would turn off the pump once somebody’s blood sugar is already high, could you get into a ketoacidosis situation? The data overwhelming says that it is an infinitesimal, if not a nearly-impossible risk, because the systems only turn off for two hours maximum.
The other thing they want to show is that the pump is not turning off all the time or frequently and you get a decrement in A1c. That is a study we think we should do. It could be fast and reasonable.
JDRF is in the midst of putting together a letter for draft guidance with help from clinicians and associations from all over the country and abroad, including the ADA, the AACE, and AADE [the American Diabetes Association, the American Association of Clinical Endocrinologists, and the American Association of Diabetes Educators, respectively] who are all submitting letters on this.
This has been very frustrating; it has already been two years since [the Minimed Paradigm Veo] has been approved worldwide, but it is still not here in the United States.

There are some meaty issues here. What this really boils down to is the risks versus the benefits, and FDA is taking a very risk adverse approach. So, you have a hypothetical diabetic ketoacidosis scenario—which is a very unlikely scenario—versus thousands of people who are getting low overnight and some of them even dying. Is that the proper balance for something that could save lives? We are arguing it is not.

DiabetesCare.net: It sounds like you are getting everyone aligned to talk with FDA to explain these studies might not be necessary and it may be taking valuable time in getting this product to market. Has the FDA responded?
Dr. Kowalski: We have a very close partnership with FDA and meet with them quarterly. The artificial pancreas project has been put on the Critical Path initiative. This is a medical initiative where areas of focus could lead to improvement in public health. It has been an active discussion. FDA has come back to us saying, ‘we do want to hear input from the clinicians.’
On another front, one of the things we have been doing with FDA is working on future guidance for insulin dosing systems—the next serious step in the development of the artificial pancreas. 
We have worked with a number of leading clinicians to submit draft guidance to FDA on this topic. [We have asked them] what would patient advocates or doctors who treat thousands of patients with diabetes like to see as a reasonable pathway to FDA approval? What would they see as important in clinical studies?

FDA has promised guidance on this by December 1. When this guidance comes out, we hope we can strike a reasonable balance with them.

This is part one of a two part interview with Dr. Kowalski. The second part of this story features Dr. Kowalski discussing the possibilities of full automation, and when people with diabetes in the U.S. can expect to see a variation of the artificial pancreas on the market.