When the JDRF recently announced its newest artificial pancreas push — a partnership with Animas and Dexcom to actually develop a commercial product — head of the project Aaron Kowalski kindly agreed to answer reader questions here.
Today, I bring you those answers, direct “from the horses’ mouth,” as it were.
Q) How much more advanced will this be than the Medtronic Paradigm Veo system launched in Europe, which stops insulin delivery for a period of time in response to a low?
A) I believe that the Veo is a very important step forward. It makes no sense to me that today you can wear a CGM, have it alarm you that you are super low, and your pump continues to pump insulin into your body. The Veo addresses this problem.
There are two main differences here. First, on the hypo side, the new system would react before hypo occurs to reduce or stop insulin delivery. Preliminary studies that we’ve done in the JDRF Artificial Pancreas Consortium show that while it will be difficult to prevent all hypo, that more than ¾ of hypo was prevented.
Second, this system would react to hyperglycemia and for the first time automatically dose insulin. We call this a “treat to target” approach because the system will not attempt to hit an aggressive glucose target, but rather it will attempt to minimize time spent at high glucose levels.
Regarding accuracy of CGM systems, will this project be using Dexcom’s existing technology or will they work to further improve its accuracy before integration?
I believe that we can use DexCom’s existing technology, but as the project evolves there is no doubt that newer technologies will be incorporated if and when they are developed. There are many people out there who will say, “I wear the DexCom and I don’t know if it’s accurate enough for a closed-loop system.” I agree that it may not be accurate enough for a fully closed-loop with a low target glucose level. But again, the idea here is a treat to target approach. This provides a buffer away from hypoglycemia.
How likely is it that the new product will be a truly integrated device with only ONE infusion set?
As much as I would love this — I think it is unlikely right away. There are more technical issues that need to be worked out to combine the pump and sensor sites. This is certainly a goal for the field and JDRF is actively looking at research that may accelerate the delivery of such a technology. The goal here is to make today’s system much more effective and safe and from there we hope we’ll rapidly see innovations that combine sites, miniaturize, integrate into cell phones, etc. — all of the things that you’ve written about on Diabetes Mine.com for the past years.
What about active insulin/ insulin on board (IOB, both carbs &/or corrections)? Will the system take into account how long it has been since eating and how much active insulin is working at any given moment?
Absolutely. Understanding the amount of insulin on board is a critical safety element that is a key part of the algorithms being developed.
What about use during exercising? A patient might still want to set a temporary basal rate. Will there be a mode on this system to disable automatic insulin adjustments?
The system will allow for full control of one’s diabetes. The aim is for it to “help” when the glucose is dropping or rising out of the target range and the person with diabetes isn’t aware. Interestingly, we’ve seen tremendous results in clinical trials that have been performed in the JDRF APP Consortium. They’ve shown that the algorithms do a fantastic job minimizing exposure to hypoglycemia post-exercise.
Any thoughts about incorporating glucagon or another substance to raise blood sugars into the pump? (simply turning off the pump to correct lows is often much too slow to raise blood sugar immediately)
We are funding two world class labs (one in Boston and one in Portland) that are studying the use of glucagon in an artificial pancreas. The initial results have been amazing. Great control is achieved. That said, there are challenges that make this research much likelier to take longer to realize than an insulin alone approach. First, as everyone here knows, glucagon currently comes as a powder. This is because glucagon is not nearly as stable as insulin and breaks down over time. So, to have an insulin/glucagon system we need new glucagon and unfortunately not much work is happening on this front and having an FDA-approved, stable, pumpable glucagon will take time.
Second, there are no dual-hormone pumps and this is not a big area of research — in fact almost nobody is working on it. As you can imagine, there are not too many people willing to wear two pumps. So, we need a new type of pump to be developed. That said, this approach is a goal of the JDRF APP and these are areas of research we are likely to support in the very near future.
Many people have highlighted that we can’t prevent all lows without glucagon or obviously with food. My feeling is: let’s take what we do today and improve upon it. While we may not be able to prevent all lows, I would argue that any lows that we automatically prevent would be a very good thing. Initial research in the Consortium suggests that many/most of overnight lows could be prevented by turning down or off insulin delivery. I think that is huge! We shouldn’t be comparing this system to a fully functioning pancreas but looking for improvements from the challenges that people with diabetes currently experience every day.
Will use require a regimented lifestyle of “feeding the machine” to calibrate and keep it humming accurately? i.e. are you going to work on shortening the calibration period of the CGM piece?
Calibration will be critical, but I don’t know that we’ll need to “feed the machine.” I think one of the interesting things that I’ve seen in the Consortium research is that the algorithms may allow for “smarter” calibration allowing for improved system performance.
How much and what percentage of the research and development money is each party contributing? Why do Animas and DexCom need JDRF’s money?
JDRF is contributing $8M to the project. The financial terms of the deal, other than JDRF’s funding, are not being disclosed. However, Animas, being part of a global top 20 pharmaceutical company, is able to bring the resources and capabilities to this project that will enable the project and the field to move forward.
JDRF plays a unique role in diabetes research, serving as both a catalyst and a direct funder of research leading to a cure and better treatments for people with type 1 diabetes and its complications. Within that role, we work with various industry participants to focus attention on diabetes, speed decision-making and collaboration, fill gaps in the therapy development pipeline, and make certain that research always remains focused on the needs of the patients with type 1 diabetes.
To ensure that research moves forward towards a cure in a directed, accelerated, and as effective a manner as possible, with the patient in mind, JDRF will often partner not just with academia but with industry participants — sometimes including those with substantial research and development resources. This partnership is an excellent example of how JDRF’s involvement, leadership and funding are driving the development of a product for people with diabetes in a far more expedited time frame than would otherwise have been the case.
Does the JDRF/Animas team feel confident they can incorporate the sophisticated algorithm into the small form factor of a pump? If not, why will it take FOUR years to get a working prototype to the FDA for review?
There are a number of time-consuming stages here. I imagine much of this will be the clinical trials required for regulatory approval. JDRF partnered with FDA over the past four years to proactively address these issues and the FDA named the Artificial Pancreas a “Critical Path Initiative” in 2007. I would argue the pieces of the puzzle exist or are close, and the goal here is to move forward as aggressively as possible.
Why did JDRF choose Animas over Medtronic? Given that about 70% of pumpers use Medtronic products, wouldn’t it be more helpful to help them develop this?
JDRF has been working with a range of device manufacturers throughout the course of our research with CGM and an artificial pancreas. Animas obviously has the technical expertise to be an integral part of this project. They have a significant presence in diabetes and a track record in bringing products and systems to market. Many people use Animas pumps today. Animas also shares JDRF’s commitment to and excitement about the prospects of developing a closed loop artificial pancreas in a relatively short timeframe, to significantly benefit people with diabetes in managing their disease on a day-to-day basis.
Our hope and belief is that additional companies will follow J&J’s lead and begin development of their own versions of an artificial pancreas system. JDRF remains interested and will continue to work towards partnering with other companies to accelerate the development of multiple diabetes management device systems.
If the proposed artificial pancreas is successful, what is JDRF’s percentage of the future profits?
If and when a system is successfully developed and commercialized, JDRF stands to realize benefits from the investment that will in turn flow directly to our future research funding capabilities. These funds are used for the organization’s charitable mission, enabling us to increase our funding of other research into cures and treatments of type 1 diabetes and its complications, as well as the ability to impact clinical usage, insurance reimbursement, and the availability of the artificial pancreas and other products and services for all people with diabetes.
Also: if and when and this project ever starts making money, will we be able to trust that JDRF’s recommendations are based on science rather than financial ties to Animas and DexCom?
The JDRF has never endorsed a specific product or company. We are and will remain a non-profit committed to curing diabetes and its complications through the support of research. We want to see multiple outstanding solutions so that people with diabetes have multiple options.
Given the other partnerships JDRF has facilitated, how soon might similar products from some of the other vendors be out?
JDRF has a number of partnerships across all areas of diabetes research and at each stage of research (see this link). Some are early stage and will take time. Others are moving to the latter stages of development.
Will you be working with insurance companies in the years prior to the release to ensure swift coverage? And how can we be confident that insurance will ever cover this when they don’t even cover many CGM systems/requests now?
Yes, we will be working proactively with payers to ensure that people with diabetes have access to these tools when they are developed.
As you know, we’ve done this with CGM and this has been a huge success story. While there are certainly some challenges out there right now, we’ve seen a dramatic improvement with most major plans now with formal positive coverage policies in place for people with type 1. JDRF was a major driver of this process — with the JDRF CGM trial providing the evidence that the payers reference as conclusively demonstrating the benefit of CGM use in glycemic control.
Will JDRF let us earmark our fundraising money to go towards a cure? For those who only want to donate to that cause?
Absolutely. We realize that each person with diabetes has different areas of research that move them and directing donations is certainly not a problem.
Thank you, Aaron. Pretty exciting stuff for us all, skeptics and supporters alike!
Note that JDRF itself loved my idea; they are hosting a live Q&A video session with Aaron Kowalski tomorrow (Wed, Jan. 27), at noon EST.
To view that event, on the JDRF’s “Promise Campaign” blog, just click here.