The Artificial Pancreas For Diabetics Is Nearly Here

the_newsbeagle writes: It's the tech that type 1 diabetics have long been waiting for: An implanted "closed-loop" system that monitors a person's blood-sugar level and adjusts injections from an insulin pump. Such a system would liberate diabetics from constant self-monitoring and give parents of diabetic children peace of mind. Thanks to improvements in glucose sensors and control algorithms, the first artificial pancreas systems are now in clinical trials.

Comments

Awesome

[MightyMartian]

Pretty amazing advance. Now I wish they'd do the same for the thyroid. My wife had hers removed due to cancer nine years ago, and has to manage her thyroid levels via synthetic thyroid hormone pills, which, while effective, are crude and require regular testing to make sure she's not hyperthyroidic or hyothyroidic.

Re:Awesome

[cdrudge]

require regular testing to make sure she's not hyperthyroidic or hyothyroidic.

How regular is regular though? Taking a test once every 3 months and adjusting your medication up or down a fraction of a milligram is an inconvenience. But it's a lot better then having to test 4-8 times a day and making significant changes to the dosage level due to what was ate, activities planned, illness, and how the body reacts to all of that. As a diabetic myself, I'd trade in an instant having to deal with hyper/hypo-thyroidism over diabetes any day.

Re:Awesome

[MightyMartian]

Except that it isn't an instant. Insulin takes effect pretty darned quickly. Thyroid changes can take days or weeks, and the synthetic hormones themselves actually have to be taken under specific circumstances, as absorption into the blood stream orally requires no significant intake of food. My wife takes her medication early in the morning and then cannot eat for something like three hours.

Having an artificial thyroid that would more closely monitor TSH (thyroid-stimulating hormone, the way your body monitors and adjusts thyroid hormone levels is complex) and adjust actual thyroid hormone levels directly would be far better in the long run.

Re:Awesome

[thegameiam]

To deal with the "no food" issue with synthroid, I think right before bed works pretty well.

Re:Awesome

[ColdWetDog]

You would need a lot more research that (I don't believe has been done) on how often you really need to adjust your thyroid. As cdrugde mentions, thyroid testing is done infrequently - on the order of months. To have an implantable system that fired off that infrequently is probably not really worth it. Dogma is that thyroid hormones don't change that much - at least a clinically noticeable values.

Most of the thyroid replacement research these days seems to be around the issue of 'minor' thyroid hormones, mostly T3. Even with this issue, we don't have good data. And T3 is produced at about 10% of the rate of T4 (the traditional thyroid hormone) so it isn't some scarce little molecule.

But if you really look closely at how much we know about things like this, it's pretty disappointing. This sort of research is slow and very expensive. Given that thyroid pills are pennies and that people do 'pretty well' on that treatment, the impetus to improve things is small.

Re:Awesome

[ArylAkamov]

Agreed, my girlfriend has graves and has decided to keep her thyroid in the slim chance that a cure will be found. That, and she's not big on the idea of ingesting radioactive iodine to destroy a part of her body, malfunctioning or not.

Security!

[chrispatch]

Hopefully not an afterthought !

Re:Security!

[Forgefather]

Being diabetic I have looked into some of these gen 1 devices that either monitor glucose or deliver insulin (not both). They are designed to network with each other but you can turn those features are off, and some don't have any networking at all.

No thanks.

[BarbaraHudson]

Even forgetting the security issue, going around with a pump and injection line connected all the time is a lot more of a pain in the ass than current methods. Also, it can't make judgements based on future activity - you might want less insulin than normal because you're about to embark an on 3-hour bike ride, which if you take your regular dose, will make you hypoglycemic, pass out, and wake up in an ambulance or the hospital (insulin efficiency increases with activity level, which is why you need less insulin when you're about to be active for any period of time).

Re:No thanks.

[Mashiki]

My sister would disagree with your assessment, and she's been a diabetic since she was 4 years old(31 years of shots now). If my great aunt was still alive, she'd be jumping for it too since she'd been a diabetic for ~70 odd years. There are plenty of methods to avoid hypoglycemia from working out, this in itself is a huge step in the current pump a needle in you every 2/4/6/8 hours that exists now.

The biggest problem is with kids and getting them to do monitoring, tests, etc., since many of them don't understand not doing something like that will kill you. It's a concept that a 4 year old can't get, no matter how simple you explain it to them.

Re:No thanks.

[h4rr4r]

Why would this device not just adjust your glucose levels during the bike ride? I am pretty sure that is exactly what it would do in that situation.

I get the feeling someone did not RTFA.

Re:No thanks.

[geekmux]

Even forgetting the security issue, going around with a pump and injection line connected all the time is a lot more of a pain in the ass than current methods. Also, it can't make judgements based on future activity - you might want less insulin than normal because you're about to embark an on 3-hour bike ride, which if you take your regular dose, will make you hypoglycemic, pass out, and wake up in an ambulance or the hospital (insulin efficiency increases with activity level, which is why you need less insulin when you're about to be active for any period of time).

I'm rather amazed that you don't believe a computer somehow cannot monitor the "future" in real-time , as was suggested when using the Dexcom continuous glucose sensor in the hardware design.

Remember we're trying to remove the human element here, which can easily be identified as the point of weakness as well as the maintainer.

Re:No thanks.

[CastrTroy]

My understanding is that there isn't a direct relation between what's being read from the sensor and what is actually in the blood. The glucose sensor just senses the amount of glucose in the blood. So if it gives a reading of x, and then gives a dose of y units of insulin to counteract, it doesn't know that you're going to start running in 1 minute which will decrease the glucose levels further than it expected to based on the amount of insulin delivered. So, now you're going to be low on glucose. The only way to do that is to add glucose to the blood. Assuming this system does this, it can bring the blood sugar back up. However, it also doesn't know about the chocolate bar you just ingested which will again add glucose to your system in the near future.

For a system like this to work, it has to make constant tiny adjustments to your insulin and glucose levels to ensure that it always remains in a certain safe zone. Scott Hanselman did a pretty good write-up a few years back. It's really kind of depressing when you look into the current state of affairs. The diabetes industry seems to be more concerned with making money than actually solving people's problems.

Re:No thanks.

[Anonymous Coward]

Of course your beta cells don't know about any of this either and they do OK. The do sense stress hormones, so if you're worried, they might pick up on the impending run. Otherwise, they make do with blood glucose levels.

Re:No thanks.

[danlip]

A normal functioning pancreas can't predict the future either. All it can do is react to what it is sensing, just like the artificial one. It may be more finely tuned, better sensors, better algorithms, etc, but none of that represents anything that couldn't be incrementally improved in the artificial pancreas.

Re:No thanks.

[sjames]

A real pancreas doesn't know you're thinking of going on a bike ride either.

Re:No thanks.

[Rich0]

Of course your beta cells don't know about any of this either and they do OK. The do sense stress hormones, so if you're worried, they might pick up on the impending run. Otherwise, they make do with blood glucose levels.

Absolutely true, but I'd be interested in the latency in the new device compared to a pancreas. Existing pumps and continuous monitors have some latency associated with them, which means they'll never do quite as well as an actual pancreas if you're suddenly going to change your rate of glucose rise/fall.

Re:No thanks.

[trout007]

The pancreas releases glucagon which causes the body to release glucose. Without this there is a real danger of this device lowering blood glucose too far.

Re:No thanks.

[BarbaraHudson]

Yes, there are methods of avoiding hypoglycemia when working out - but they're reactive, not anticipatory. Until they also include a sugar pump, forget it.

Re:No thanks.

[BarbaraHudson]

The pump can't add glucose. Your body has a limited store of glycogen in the liver that it can draw on, after which you need calories quickly.

Re:No thanks.

[BarbaraHudson]

A normal functioning pancreas won't ever put you into insulin shock either. Type 1 diabetics don't have a normally functioning pancreas, or they wouldn't be Type 1 diabetics, duh!

Re:No thanks.

[BarbaraHudson]

No, but it has real-time measuring, not a 10 - minute+ delay. Also it's much finer grained in it's operation than the device.

Re:No thanks.

[BarbaraHudson]

The thing is, the needle pricks are no big deal any more. They used to hurt a lot more than now, but the finger stickers are sharper, the amount of blood you need is next to nothing (and can be drawn from places other than the finger), and the injections are pretty much painless. And now they're experimenting with inhaled insulin.

Re:No thanks.

[Anonymous Coward]

Of course your beta cells don't know about any of this either and they do OK. The do sense stress hormones, so if you're worried, they might pick up on the impending run. Otherwise, they make do with blood glucose levels.

type 1 diabetics do not have functioning beta cells that is the general problem. The raising of blood sugar, the action of glucagon is not a problem for some type 1 diabetics. Some type 1 diabetics, myself included still have "Alpha"cells that produce glucagon by the pancreas, from there , raising of blood sugar is a task performed by the adrenal glands and the liver and the muscles (adrenal glands produce cortisol and the liver and pancreas store glycogen that on demand can be converted to glucose and released into the blood, raising blood sugar.)

stress hormones only come into play in a minor way, they are not as influential as insulin, exercise and glycogen in determining blood sugar behavior.

Re:No thanks.

[tirerim]

Um, insulin pumps have been around for about 30 years, and very common for the past 15. They're already a huge improvement over injections. This level of "artificial pancreas", though, not so much. The glucose sensing technology, though dramatically improved from its debut a decade ago, is still primitive: it uses interstitial glucose, and lags behind actual blood glucose, requiring regular calibration with fingersticks. Combine that lag with the fact that a non-diabetic pancreas starts producing insulin even before food hits the bloodstream, and it's impossible for the system to react to food in a timely way. Worse than that, though, is the fact that insulin effectiveness isn't constant. It's less effective when you're eating a lot of fat, and it's more effective when you're exercising (and for some time after). Just measuring glucose levels isn't enough to tell the system how to react. Maybe eventually they'll manage to combine this with enough other sensors to actually be a real artificial pancreas, but I think that's a lot farther off, and we'll probably have effective islet cell transplants without immune rejection by then.

Re:No thanks.

[Mashiki]

So is the pancreas, it's a reactive system to external demands as well. So I guess it needs a sugar pump...no wait...it either dampens or allows glucose production by the liver as needed. Seriously, do you know how this stuff works? I suppose that could be a problem of course if you don't have a liver...but then again, if you don't have a liver you're going to have more serious problems anyway.

Re:No thanks.

[Mashiki]

You can have insulin shock with a normally functioning pancreas, it can happen from a variety of things, including working out and sudden shocks to your system from stress, traumatic events, and so on. Saying it can never happen is assine.

Re:No thanks.

[dens]

Scott is right, I have been hearing about a cure since 1978. There have been 2 big breakthroughs in that time: the insulin pump and the CGM. Both of these have made my ability to control my blood sugars in immeasurable ways.

What I would really love to know is why I (ok, my insurance) am still paying roughly the same amount for diabetes supplies as I did 30 years ago: test strips (~$1 each, I test 5-8 times/day), pump reservoirs, infusion sets, glucose sensors ($95 for a 6 day sensor) and let's not forget the $5,000 pump that last 5 years on average, so let's say $1k/year. This is an expensive hobby. It's worth every penny to me, and even after insurance, I average $3-4k/year out of pocket. I wonder how someone who make half what I make pays for this.

My guess is that this cost is largely profit for the manufacturers. After 30+ years on the market, the strips are probably 1 penny or less to produce. There's lot of money to be made perpetuating maintenance of diseases.

Re:No thanks.

[dens]

Ok, it's late.

Both of these have made my ability to control my blood sugars in immeasurable ways.

should be:

Both of these have improved my ability to control my blood sugars in immeasurable ways.

Will Slashdot ever get post editing ability? It's a shame there aren't any geeks here to help make this a reality. ;-)

Re:No thanks.

[sjames]

The feedback loop will be longer than with a natural pancreas, so the control won't be as tight, but I'm not so sure it will be that loose. In any case, manual management by injecting a bolus of longer but slower acting insulin analogs isn't going to be as good as a natural pancreas either. The pump will be doing continuous injection of faster and shorter acting insulin.

If that proves inadequate, they could always add a couple demand buttons, one to indicate the patient has started eating (the natural pancreas does get signals for that) and another to indicate exercise is beginning. Still not perfect, but it has real potential to provide much better control than manual injection.

Not a pancreas

[jklovanc]

Since it does not produce it's own insulin it is not an artificial pancreas. It is an automatic insulin pump. It still has to be refilled with insulin periodically. It is an improvement but is not a replacement for a pancreas.

Re:Not a pancreas

[nordee]

You're being pedantic. You could make the same argument about an artificial heart. "But it requires an outside power source! It's not REALLY a replacement for the heart."

Yes, there's some marketing going on, but the article clearly states that the advantage of this system is that it works without user input. That may not seem like a big deal to you, but as the father of a Type 1 Diabeted I can assure you that it's actually a very significant change.

Re:Not a pancreas

[jklovanc]

The pancreas produces several hormones. The automatic insulin pump deals with only one. Therefore an automatic insulin pump is not a replacement for a pancreas. The heart moves blood and that is all. An artificial heart replaces that function. The other think is that artificial hearts are not a permanent solution. They get replaced with organic hearts.

That may not seem like a big deal to you,

I never said it was not a big deal.

It is an improvement but is not a replacement for a pancreas.

Re:Not a pancreas

[Chris Mattern]

The pancreas produces several hormones. The automatic insulin pump deals with only one. Therefore an automatic insulin pump is not a replacement for a pancreas

True in the pedantic sense, but it doesn't need to be one. The problem in juvenile diabetes isn't that the whole pancreas shuts down. Only the parts that make insulin ("the Isles of Langerhans") do. The rest of the pancreas remains perfectly functional. I suppose that they should have really called it an "artifical Isles of Langerhans", but that's pretty clumsy.

Re:Not a pancreas

[jklovanc]

I suppose that they should have really called it an "artifical Isles of Langerhans", but that's pretty clumsy.

Sorry but still incorrect. The Islets of Langerhans produce all the hormones and not just insulin. Even the specific type of Islets that produces insulin produces another hormone.

Beta cells producing insulin and amylin

What you call pedantic others call accurate. It is unnecessary to hype a significant advancement by over stating what it is.

Re:Not a pancreas

[Livius]

I would agree that "pancreas" is an exaggeration but it's really good news nonetheless.

Re:Not a pancreas

[Skapare]

and you can't store too much ... body temperature, time, and all that. production as needed would be better.

But is it hackable

[penguinoid]

Just wondering, can it be modified to work with caffeine?

Re:Only now?

[Forgefather]

Most test strips are one time use. The big issue here isn't that they can monitor the levels, but that they can do it accurately. Even now there is a variance of about 10% in the current models. Accuracy also goes down over usage and the last thing you want is to send some poor sap into hypoglycemia because the pump got a wrong reading.

It's not an artificial pancreas

[Anonymous Coward]

It's a self regulating insulin pump. That's a wonderful for type 1 diabetics, but the pancreas produces more than just insulin.

Glucagon is the primary axis hormone to insulin. A true artificial pancreas would monitor both hormones to optimize that relationship.

Gilding the lily diminishes it.

[140Mandak262Jamuna]

Unless the device makes insulin on demand, it ain't artificial pancreas. It releases insulin stored in a tank, mimicking the action of the pancreas, but the tank has to be refilled from external sources of insulin. But still it is a great advance. Why do they have to ruin a good article by needlessly hyping it in the title?

Insulin Resistance

[briancox2]

This is an engineering breakthrough. But we are still waiting on a medical breakthrough.

The real problem in diabetes is not limited to having the correct amount of insulin in the blood in real time. The problem that makes diabetes so hard to treat is that a person's cells develop insulin resistance, requiring larger and larger doses of insulin to break down sugars correctly.

Re:Insulin Resistance

[Chris Mattern]

Well, actually, there are two kinds of diabetes. In one type (adult-onset diabetes or Type 2 diabetes) the body becomes less and less responsive to insulin. This is the kind associated with obesity, and the pump won't help this much. The other type (juvenile diabetes or Type 1 diabetes) is caused by the body stopping production of insulin, generally because the insulin-producing cells of the pancreas die off. The body remains completely responsive to insulin; the problem is that there isn't any any more. This will be a godsend for people with Type I.

Re:Insulin Resistance

[Anonymous Coward]

You don't have Type I Diabetes.

Insulin resistance is a Type 2 Diabetes issue.

Re:Insulin Resistance

[russotto]

They are at their furthest apart still symptoms of the same lifestyle problems.

No, Type 1 diabetes is an autoimmune disease, not a symptom of a lifestyle problem. It's not caused by lifestyle either.

Re:Insulin Resistance

[dens]

Same here. I have had type 1 diabetes for 37 years and an insulin pump for 29 years. I also use a CGM. My insulin sensitivity hasn't changed in at least 30 years, except temporarily due to other factors, e.g. if I have a cold or some other infection that makes me less sensitive to insulin (once my sugar was out of whack big time for 2 weeks after a cortisone shot in my shoulder)..

My problem is with low blood sugars. I play a lot of sports and the trouble is that this causes my insulin to affect my blood sugar much more than normally and it isn't something a CGM can properly handle yet, especially with the lag. If I play basketball, the 20+ minute lag of a CGM can men that it thinks my blood sugar is 140 when it's really 50 (way too low). In addition, after shooting hoops or lifting weights, my insulin sensitivity can be affected for up to 48 hours after that.

Re:Insulin Resistance

[Anonymous Coward]

I did just recently have a doctor lecture me that they are no longer differentiating as strongly between type 1 and type 2 diabetes in their treatment approaches. They are at their furthest apart still symptoms of the same lifestyle problems.

I would not trust a doctor that believes that type 1 (pancreas not producing the insulin hormone because its cells die from an autoimmune reaction, often at a young age) should be treated in the same way as type 2 (where the pancreas has to make extra insulin, because the cells becomes less responsive to it)

Re:Let's all hope

[Dunbal]

You must be new here.

Re:$ type 2 ?

[wcrowe]

Cute. Yes, lifestyle changes make a difference for about 20% of type 2 diabetics, but for the rest, treating the disease is elusive. There are a lot of type 2 diabetics who are not overweight and follow healthy eating habits. There are a lot of obese people who never develop type 2 diabetes.
 

Re:$ type 2 ?

[butchersong]

I would be interested in seeing the details the lifestyle changes tied to those numbers if you happen to have them. Everyone I've spoken with has had tremendous success with change of diet and exercise. Maybe this was still incorporating the standard american diet with only moderate changes?

Hope robopancreas is secure...

[Anonymous Coward]

Here's hoping that whoever brings "robopancreas" to market pays very, very serious attention to security. Security right now in medical devices is woefully poor and the medical device industry really needs to get their act together. There are evil folks out there who create malware that target insulin and morphine pumps...

Re:$ type 2 ?

[wcrowe]

I was quoting the percentage off the top of my head. The number is actually lower. Here is an article on the CDC study that the numbers come from:

http://www.livescience.com/256...

I've been living with diabetes for over 12 years now. At one point I had lost 110 pounds through diet and exercise. At first my situation improved, but over time it got worse. I now live life as a type 1 diabetic.

You can certainly live better with diabetes through lifestyle changes, I'm not going to argue that. But for most type 2s, treatment will eventually still include medications and even insulin injections.

Cost

[wlowe84]

Affordability will be an issue for most. My wife used to use an insulin pump, the pump cost around five grand. The consumables necessary to keep it operating month after month, year after year add up quickly. The initial cost of the device is nothing compared to the cost of maintaining it. After a few years it became impossibly expensive for us to continue to use. Back to syringes and vials for us. Pumps can work great, but if it takes the house and car payment to use them, quality of life can become an issue.