Carb DM Bay Area Diabetes Summit: Expert Panel on Automated Insulin Delivery

By Helen Gao, Adam Brown, and Alexander Wolf

At the recent Carb DM Bay Area Diabetes Summit – a local favorite of ours! – three experts on automated insulin delivery (aka, “artificial pancreas,” “AP,” or “closed loop” systems) discussed the current state of the field, what patients can expect with “first-generation” systems, working with the FDA, and future innovation. The all star panel included leading closed-loop researcher Dr. Bruce Buckingham (Stanford), Medtronic Diabetes’ Chief Medical Officer Dr. Fran Kaufman, Bigfoot Biomedical’s Director of Clinical Innovation Jen Block, and moderator Howard Look (Tidepool). See the highlights below!

What were the key takeaways from the panel?

1) Dr. Kaufman highlighted that the pivotal trial for the MiniMed 670G hybrid closed loop system has completed (see our past coverage), and shared that Medtronic hopes to provide the 670G at a relatively comparable price to the cost of a separate CGM and pump system – that would be amazing news for patients if it is possible. Dr. Kaufman was honest in what she is most worried about, however: “The biggest threat to this whole field is who’s going to pay for it and what’s the cost going to be.” We’re not sure if Medtronic will have an upgrade program for current pumpers, though we hope some discount is possible for those already on a pump and still in warranty. Dr. Kaufman also shared that Medtronic has spent tens of millions of dollars on trials for the 670G alone, and the official FDA submission report will be tens of thousands of pages long – it’s a big investment to test a system that automates insulin delivery! [Editor’s Note: The 670G will be submitted to the FDA before the end of June, and the hope is a US launch by April 2017.]

2) Dr. Buckingham said the FDA has “accelerated tremendously” on automated insulin delivery, which he attributed to the power of the patient advocacy community (hear! hear!). One area of FDA uncertainty, however, is the approval process for next-generation improvements: what can companies update on their own without a long FDA process, and what will require more in depth review? This is particularly important for automated systems that connect to the cloud and smartphone apps, which could be rapidly improved with remote software updates. Noted Dr. Kaufman, “If every time we [artificial pancreas developers] want to innovate, we have to go through an arduous FDA process, we’re all going to get stuck.” On the other hand, Dr. Kaufman noted that the FDA is already psyched about future innovation: “The FDA has said repeatedly, ‘This is gen one and you better have gen two, three, and four in mind’ because they want to see innovation.”

3) Jen Block confirmed that the Bigfoot still plans to begin a pivotal trial of its automated insulin delivery system in 2017. The hope is to submit the device to the FDA by the end of 2017, with potential for a commercial launch in 2018. See our previous coverage of the interesting design, which will use a smartphone app as the user interface (no pump screen or buttons).

PANEL DISCUSSION: Realities of Bringing Automated Insulin Delivery to Market

Mr. Look: The theme I hear is “wow, this is really hard, but there’s very good evidence that being on a closed loop system totally works.” My question is, why does it take 100,000 pages or until 2018? What is the thing you would change that would allow closed loop systems to get in people’s hands as quickly as possible?

Dr. Kaufman (Medtronic): It is a process. There is a need to show safety first and efficacy – does it really improve somebody’s outcome? We have had a series of FDA meetings monthly for over a year. This has been priority for a lot of us. We would rather send a short memo than so many pages of documents, but I don’t disagree that we need a regulatory process. It’s been a good journey and we’re kind of there. I would hate to live in the US without any regulatory processes. It could be a little easier – if they could speed read very quickly, that would be nice – but we’re almost there. As the first company, we’ve taken the brunt of a lot of unknowns, as we should. The process will also push everyone to keep innovating. The FDA has said repeatedly “this is gen one and you better have gen two, three, and four in mind” because they want to see innovation.

Ms. Block (Bigfoot Biomedical): I agree with you, the process is really important. Many of you will be using these systems someday and the burden on us is to demonstrate that they’re both safe and effective. My biggest hope is that the process could be a little faster. On the outside looking in, there’s a strong desire to push further and go faster. On the inside, I see that the questions are reasonable and that they take time to show. All of these things take time; my biggest hope is that it would just be faster.

Dr. Buckingham (Stanford): The reason it’s moving so fast at all is the patient advocacy that pushed hard on the government. The government can move slowly, but it can be moved. The process has gone so much faster – it’s accelerated tremendously in the last two years.

Q: How do you keep glucagon from going bad in the bi-hormonal devices?

Dr. Buckingham: There are a couple companies out there working on stable glucagon. One of them is Xeris, which puts in something called DMSO to make it last for more than a year. However, it will freeze in the fridge. There’s also a Danish company called Zealand that has modified its glucagon to be stable in liquid form but with the same biological action. [Editor’s Note: The Bionic Pancreas team has talked about using one of these glucagons in its pivotal study of the iLet.]

Mr. Look: There are some companies like Medtronic and Bigfoot doing a single hormone system. Now some companies are doing bi-hormonal as well. Can you comment on the advantages and disadvantages of both?

Ms. Block: Dr. Buckingham pointed to research comparing insulin-only and insulin plus glucagon systems this morning. Insulin is well known and we have lots of experience delivering it, so it’s nice to start there.

Dr. Kaufman: It was a conscious decision to not go to glucagon at this point. Nobody has ever gotten glucagon every day in these quantities. It adds a complexity that we don’t feel is warranted – data shows that at this point, there’s not much of a difference between dual and single hormone. Move all that aside and just talk about cost – the 40%, 100% increase in cost, whatever it is. In my experience with the 530G, even with no price change whatsoever, a number of insurance companies said they didn’t feel comfortable covering it. What’s the added benefit with the added cost? We’re hoping there won’t be a significant difference between the cost of the 670G and what our sensor and pump cost now, even though we’re spending millions in clinical trials. We need to be aware of cost.

Dr. Buckingham: The fact that this is going to cost about the same is music to everyone’s ears. That said, glucagon is potentially a benefit in exercise situations. Also, it might be a benefit for someone with severe hypoglycemia unawareness, who maybe needs a glucagon pump.

Ms. Block: Until we have more experience with use, glucagon may be better for next-generation systems. We talked about some of the challenges of using a mobile app for the system and making upgrades to a system, but if you use a mobile app, you do have flexibility to do updates over the air. [Editor’s Note: The Bionic Pancreas team is working on both insulin-only and insulin+glucagon versions of its system. This will be tested in a pivotal trial in 2017.]