Breakthrough Diabetes Tech, But Don’t Call it an Artificial Pancreas

By Gary Kaye,

Chief Content Officer, Tech 50+, Technology Through Our Eyes

Diabetes sucks.  Why mince words?  The disease has dire consequences.  It’s the disease that keeps on giving.  It gives you kidney disease.  It gives you retinopathy.  It gives you neuropathy.  It gives you non-healing wounds and amputated limbs.  It can devastate your other organs and circulatory system. It can be incredibly difficult to monitor and control.  And more than 29 million Americans have it.  According to one recent study between 1/3 and 1/2 of U.S. adults are either diabetic or pre-diabetic.  And not only is the cost to the sufferers horrendous, but the impact on the health care system runs over $250 billion each year.

It’s no wonder that diabetics have been closely watching the development of biomedical devices that will help them monitor and control diabetes. Most need to prick their fingers four or more times a day to measure their blood glucose.  Some can control the disease with pills, others need insulin injections.  And often they are plagued with huge blood sugar swings from highs to lows which themselves can be dangerous, or deadly.

Now, for the first time, the FDA has approved what’s called a “closed loop system” which continuously monitors a patient’s blood glucose and delivers micro-doses of insulin to keep that level in check.  The system has also been referred to as an “artificial pancreas,” a phrase that does not sit well with the American Diabetes Association since the device only performs one of the many functions of that organ.

The two major physical components of the system, the blood glucose monitor, and the insulin pump have each been around for years.  Dean Kamen invented the insulin pump in 1973 and first began marketing it in 1976.  Dexcom first brought continuous blood glucose monitoring to market in 2006.  But only now are the two systems being combined to work together. This has not been an easy problem to solve, and it’s taken sophisticated algorithms to make it work right.  That’s because insulin takes a while to get into the bloodstream while consuming foods can have an almost instant impact on blood sugar levels, so you are always playing catch up.

Medtronic MiniMed 670G

The first FDA approved closed loop device is the Medtronic MiniMed 670G.  So far the device is only approved for Type I diabetics, though they constitute only about 10% of the nation’s diabetics.  Type I diabetics generally require what’s called full insulin replacement, in other words, their bodies simply don’t make insulin.  Only a fraction of Type II diabetics are in the same boat – many more can keep it in control by exercise, diet, and medication.

We spoke with Mike Hill, Vice President for Global Marketing at Medtronic who says the MiniMed 670G is a hybrid closed loop system, but he says that while it’s a big step forward, it’s still not the goal,

From our point of view, a full closed loop system would be one that is constantly monitoring glucose, and then the system is automatically adjusting and delivering basal and bolus insulin so that it wouldn’t require any patient interaction beyond wearing the devices. That would be a full closed loop. The mini-med 670 G is called a hybrid closed loop because the pump does some of the automation but the system does require patient interaction. Our current hybrid closed loop system has a continuous glucose monitor, so it’s checking glucose readings basically every 5 minutes and then the system will automate the basal rates, the background insulin. It’ll take that up or down based on what your glucose level is with a target of 120.

In other words, says Hill, the patient still needs to intervene by giving himself a bolus through the pump, generally around mealtime, and periodically taking manual blood glucose measurements to keep the system calibrated.

As to the nickname “artificial pancreas,”

We don’t talk about an artificial pancreas or path to an artificial pancreas. We used to several years ago use that terminology and that’s why for example you’ll still see the FDA and other people using it but we received feedback from clinicians, patients, and other folks that a pancreas does more than just deliver insulin, or monitor glucose, which from Medtronic’s point of view, is what our systems do, so the implication was more than it could possibly be. We talk about a closed loop, basically closing the loop between measuring glucose and delivering insulin.

There are two different families of insulins: basal, which is the long acting background insulin sold under names like Levemir and Lantus,  and bolus, a short-acting insulin such as Novolog and Humalog. Hill explains,

So what we call basal insulin is the background insulin. It’s the low steady rate that allows you to bring basic glucose into the system. Bolus is when you need a specific larger amount of insulin to cover a specific event. So you eat a meal, as your body digests the food you’re going to get a surge of glucose in your blood as your intestines are absorbing it. If you don’t have extra insulin in your body times with when that food is digested, you’re going to have really high glucose levels for a long time. That can have both short and long-term complications. High sugar levels in the blood are damaging to the body.

Hill says a major advantage of the closed loop system is that it automates what has been a very complex process,

You’re constantly making all these decisions every day about what you’re eating, are you exercising, are you sick or not, how much insulin did you take before, when are you going to bed, and you’re trying to put all that info together and then make decisions about what you should do with insulin. That kind of complex repetitive mass based decision making is not what humans are great at. We do it. But it’s frustrating and challenging. So the benefit of starting to automate these things is because the system is checking glucose every 5 minutes and then the system is making a decision every 5 minutes, it can automate a ton of small decisions that will lead you to have a better outcome.  In that example you could put a small amount of insulin on board and eat, not whatever you wanted, but choose a different type of meal, and if you put in too much or too little insulin, the system is going to see where your blood sugars are going and then dial up or down the insulin to match. Basically, it tightens up every decision. The only way for you to replicate that yourself is taking finger sticks every 5 minutes and giving little bits of insulin and that sort of thing which is far more maintenance than you’d want to do. So, in short, the system can give you better control with less burden so people are getting both clinical benefits and quality of life benefits.

Beta Bionics iLet

While Medtronic is the first to bring a closed loop system to market, there are a number of other companies who are lined up for the FDA approval process and are expected to have their own devices commercially available over the next several years.  The next one up may well be the iLet from Beta Bionics, which began as a joint project by Boston University and Massachusetts General Hospital.  I spoke with Beta Bionics CEO Edward Damiano who says that in the near-term his company plans to produce two different devices, one, which like the Medtronic device, only controls insulin levels.  But Beta Bionics is planning a second, which he calls a bi-hormonal device, capable of delivering both insulin and glucogen.  Such a system could better control low blood sugar levels.  He anticipates starting trials later this year and hopes to have at least the first device on the market by 2019.  Damiano says there are significant differences between his iLet and the MiniMed 670G.  The iLet will be more autonomous and require less human intervention.  He puts it this way, “if the Medtronic device is like putting cruise control into a car, ours is like having a self-driving car.”  The iLet is initially expected to use the Dexcom sensor.  But he anticipates that the sensor technology will continue to improve, as will the algorithms; the computer enabled mathematical calculations that will continuously monitor, anticipate, and adjust how the system is operating.

Doctor Stuart Alan Weinzimer is a pediatric endocrinologist who is currently supervising the Medtronic closed loop clinical trial at Yale-New Haven Medical Center.  He says the early results are very positive.

In the initial pivotal study we rolled 124 people, 94 adults, 30 adolescents, it was not a randomized controlled study, everybody got the system. After a 3-month period, with very significant reductions in A1C and concomitant reductions in hypoglycemia (low blood sugar) exposure which is really the best of both worlds…In the old days you had to lower A1C if you increased your risk of hypoglycemia but here we seem to achieve both. After 12 thousand patient days of use there weren’t any episodes of severe hypoglycemia or DKA. It really was quite safe. Most people who are familiar with these devices know it’s not a completely automated system, it’s a hybrid system. For meals, you still have to count your carbs but then the system will have a variable that will work based off its sensors. In between meals and overnight, it’s really good at regulating despite whatever would happen at night, whatever you would eat or whatever exercise you did during the day, the nighttime were really quite good. People would tend to wake up in the morning pretty close to goal blood sugar. Starting off the day at 140, you don’t lose half the day trying to get there.

Dr. Weinzimer also predicts the systems will get even better as sensor technology improves.  But he is concerned not only about making them better, but also making them affordable,

I actually am concerned about the insurance reimbursements because people aren’t going to be able to afford these devices if they’re not covered by insurance. No one’s going to be able to layout $10,000 and all the consumables and the sensors unless it’s covered and you’re getting reimbursed for it, but I have patients who are on these devices that were not in the study. It’s about 6 weeks now. 4-6 weeks now that I’m starting to get regular clinic patients get on this device.

While some insurance companies have started paying for them, there’s still a major issue as to when Medicare will get on board as well, and that’s going to make a huge difference in the adoption of the systems.

Right now, the Medtronic device is approved for Type I diabetics ages 14 and up.  The company’s first goal is to get that age down to seven.  After that is the battle to get it approved for the many Type II diabetics who are on full insulin replacement.  But that, according to both Medtronic and the clinicians we spoke with, could still take years.

We asked another endocrinologist, Doctor Stephen Ponder, about why there seems to be less interest on the part of the FDA in approving these devices for Type 2 diabetics even though they are the huge majority,

I think people with Type 2 are, well most of them are not taking insulin therapy and so you don’t have a strong motivation for blood glucose monitoring like you would with people who are taking insulin. Having said that, when you have a Type 2 patient who does reach the level of insulin dependency, I would think you would have just the same need as you would for a Type 1. However, one other caveat here is your average Type 2 diabetes patient is somewhat protected against hypoglycemia compared to a Type 1 because a majority of Type 2’s have a significant insulin resistance. There’s less of a concern about the low blood sugars that those of us with Type 1 have to live with on a daily basis. Having said that. There are 9 times the people with Type 2 as there are with Type 1. Again, most of them are not all at the stage of insulin dependency. They’re taking oral agents, or managing with lifestyle or lifestyle and oral agents or some of the injectable non-insulin agents out there. If they survive long enough, your average person with Type 2 does eventually require insulin therapy, that’s just the natural history of the disease. It’s progressive.

But Doctor Ponder is also concerned about both the limitations and the complexity of these first devices saying they still require a lot of human intervention, and not every potential patient can deal with that,

Insulin can work only so fast. Sensors can only sense so quickly so while you can tie the two together with software systems and all that, they also have to be maintained by a human being who needs to have some capacity to troubleshoot and problem solve those things and right now I see a tremendous amount in difficulty out there in getting people trained well enough to be able to maintain them. The studies and the trials you see, are typically in rolling patients who you and I would consider as the cream of the crop. People who are very capable, very empowered, very motivated to do all this, where as if you apply one of these devices to what I would call a free range patient, that might be different. When that happens, it’s going to be interesting. I’m experienced enough in a jaded cynical way, whatever you want to say, after doing this for all these decades that I take everything with a grain of salt when a representative of a company tells me that devices are working so well in these trials and so on. Well trials are generally done with selected patients. There’s eligibility criteria. There’s all sorts of things that go into getting enrolled in a controlled trial. I’m not surprised you can get stellar results with great patients. What’s going to happen when that hits the road with the average patient, the typical patient, as I said the free range patient. That’s what I’m waiting to see.

Ponder also feels the devices will get better as the sensors and algorithms improve, and as they evolve to require less human intervention.  Meanwhile, millions of diabetics are sticking their fingers, injecting themselves, and playing a life and death game of “catch me if you can” with their blood sugar.  So it’s not surprising that there’s a huge community of diabetics waiting with anticipation that at long last there will be an easier way to manage this often all-consuming disease.  And watching with frustration as these dramatic technologies seem to remain just out of reach for most of them.