Artificial Pancreas by 2015?

Several groups are testing ideas for an artificial pancreas to help Type 1 diabetics. For a Boston University engineer whose son is diabetic, progress is on a personal deadline.

Damiano is trying to build an artificial pancreas, similar to the model here, that could help people with Type 1 diabetes maintain healthy levels of glucose in the blood by the time his son leaves for college. (Myung J. Chun / Los Angeles Times)

Around the world, roughly a dozen groups are testing similar technologies. Damiano's team, further along than most, began testing its computer algorithm in humans in 2008 and is in its second round of in-patient human trials. Damiano says he hopes to be performing out-patient trials by 2012 and estimates that the device could be on the market by 2015.,0,1240573.story

Something like this would be amazing, even if it still required some daily checking and minor highs and lows. If it could take the 24/7 thinking/adjusting out of Type 1 it would be progress!

Phase III testing to test new insulin, CGMs and the AP algorithm are beginning at Stanford thanks to the support of JDRF and the three companies who have been working on the hardware and software to make it all work. The system has been tested in Phase II tests and now is about to start Phase III testing, i.e., real world testing.

It’s closer than you think and the meeting at the FDA on Nov 10, 2010, is being held to see what they can do to get the research on track (focused efforts) and to get advice on how it can come to market ASAP.

FDA AP hearing info:

Given the malfunctions of pumps & the inaccuracy of CGMS, I wouldn’t trust an AP. The cost would be staggering, which leaves out the majority of the world’s diabetics. I wish the millions being poured into developing this device would instead be used for reserach. My hopes are pinned on Smart Insulin, which if the forecasts are correct, would work to keep BG normal without a machine.

This would be wonderful to have while she is at college. I am even more hopeful that the doctors at Columbia will be well on their way to discovering the cure for reversing autoimmunity in Type 1; needing then only a way to replace the beta cells.

I would be even more thrilled with SmartInsulin. I know this is still in the research stage. Is there a timeframe on when this insulin may be ready for release?

I was a test subject in Damiano’s most recent study at MGH. I was on the AP for a weekend in Oct and go back for the second in Dec. Overall, it did better than I expected, but it won’t meet many Tu users’ thresholds for good control. I never went low, but my bg was averaged at ~140. I have a few thoughts:

  • The pump was bihormonal, insulin + glucagon in separate pods. Glucagon is important because insulin withdrawal is not fast enough to stop laws because it hangs around for so long once infused. I never went <70 (bg measured by glucoscout every 15 min) at any point, including exercise. I want me a pump with glucagon!

  • There will need to be some user input. In this case, it gave a dose based on body weight at the time of the meal and then responded to rises in bg. As expected, I went above 200 at almost every meal (70g cho). I was elevated for some time afterwards. If I were to get one of these to take home, I’d want a lot more user input. If I can keep most of my spikes <150 then I won’t settle for that.

  • It performed best overnight, where I was completely stable.

  • I trusted the Navigator, which surprised me. It outperformed both Dex 7 and MM Guardian during this study, and beat my expectations based on my time on the MM 722 and a short time on the Revel. I’d keep a watchful eye on it, but I was completely blown away at how accurate it was.

I would want to keep control over meals, but I would be comfortable letting it do the rest. I’d watch it like a hawk but I trusted it more than I thought that I would. Next gen insulins may be a path around the meal issue but I don’t believe that we’re there yet. The FDA and the companies will probably be more comfortable with letting me go >200 rather than risking a low, but I wouldn’t be. Without the ability to more aggressively handle meals, I’d have to politely decline for the time being.

That was a long winded way of saying that I see a lot of promise. In the short term for people unable/unwilling to go for more aggressive control and in the long term for all of us!

Thanks for the info. Just to be able to hold steady all night and never go below 70 dujring exercise would solve most of our problems. Sheer bliss. I agree I would like user input during the day for lower postprandials. How much user input is currently allowed using the AP? The main thing is this device may be able to keep children and adults with Type 1 safe from hypoglycemia.

I actually met Ed Damiano at the Annual JDRF meeting in Connecticut in 2009. He invited me to participate in his study. I went to Boston in September 2009 for tests. I unfortunately missed one of the criteria by only a 0.10%. Ironically my blood tests 3 weeks later were perfect. Must have been the oysters for dinner the night before at the Union Oyster House. hahaha. I was really looking forward to going thru his study, They were looking forward to it also since I was over the 50 year mark. Maybe now that they are in phase 3 the criteria has lessened.

I agree I think they should put more money into getting the transplanted islet cells in the liver to work. You are right…there should be some sort of “Smart” insulin…

Agree 100%. We’ve all been there where the CGM says you are 55 and you test and your are 100.

I agree that smart insulin would be better and cheaper, and think we need to keep moving forward on many different fronts, but an AP is probably a lot closer to a clinical reality than anything else. All I can find from places like SmartCell are vague cartoons on how smart insulin might work, but none of the hard science behind it. Dr Zion has no scientific articles on the topic (or much of anything else), despite being funded by the NIH and JDRF, which raises some doubts for me. I think it’s too early to bet too heavily on that horse.

The cost and accuracy are definitely going to be issues, but I see them resolving in the short term. An AP will be more expensive than a regular pump, but the hardware for now is mostly the same (except maybe for the glucagon in bihormonal versions). I think that competition, the fact that they’ll still be making their money off of consumables, and the fact that insurers will have to be willing to pay will keep them from getting too high though. The accuracy I think will also improve greatly over the next few years. If you believe the claims, cgm will be on a whole new playing field within a year. But I’ve learned not to believe too much of what I hear!

Hey Jan. This one only allowed them to say that I started eating and it gave a standard dose of a few units. They are uncertain when they build the device (the next step) what will be allowed. It almost sounded like there would be several levels based on cho content, but hopefully it will be flexible!

I agree with Tom. We need research to move forward on as many fronts as possible. Even if the AP doesn’t work out, the data on CGM and the therapeutic use of glucagon are valuable. At the very least, maybe we can look forward to more accuracy out of our CGM.

I don’t particularly care where the next breakthrough comes from. The nature of breakthroughs is that treatment options and therapies are advanced. We ultimately want a cure, but i would never want all of our eggs placed in one basket. The more funding for as many options as possible until a cure is found, the better…

Thanks for all the replies, they have been very interesting.

Tom T, I appreciate your input, you have given all of us some useful information. Ray, I wish you had been able to participate, Then you could have reported back to us. It would have been interesting.

I first saw this link on the site. There is a parent there who hopes the AP will be there for her child by the time he/she leaves home to go to college.

Transplanted islet cells do work. But who wants to take immunosuppressant drugs his whole life to prevent rejection? Or the other alternate: use stem cells from your own body so you do not need immunosuppressants. All very nice but who wants to be treated with highly manipulated living cells? Long term effects, cancer rates and so forth - this looks to me as if the risks of these current and potential treatments exceed the current risks accociated with diabetes management by far.

Tom, you said your average BG over the course of the test was about 140. That is a HbA1c of about 6.5 (depending on whose conversion equation you use). This is better than most T1’s are able to achieve, period. And it sounds like it was done without any diet or exercise restrictions and without any carb counting. Sign me up!!

How was the glucagon handled - was it loaded in the pump the same as insulin, and how often did it have to be reloaded?

Anyone know who is doing the testing in Boston? I looked at one point but couldn’t find it.

Most of the current transplantation work being done now is with encapsulated islets (usually from animals e.g. pigs) so that immuno-suppressants are not needed avoiding the immuno-suppressant risks. But although some promising work is being done, the transplantations are very expensive, and encapsulated transplanted cells don’t live all that long so it needs to be done periodically, and of course there are the xeno-transplantation risks that have to be considered.

I was off by one year. I was there in 2008 not 2009. Here is what they sent me.
5957-ADVERTISEMENT.doc (33 KB)

Meh, they can’t get a non-invasive glucose meter to work good enough to replace finger pokes. Spending billions of dollars on something like that is a big waste of money IMO. It may be helpful for a small percentage of real brittle diabetics but I can’t see it being the standard in treatment. We should here some real critical news about Exsulin’s human trial in the next few months and hopefully its positive. Obviously I’m eager to see the Smart Insulin trials start.