Automatic "artificial pancreas" for diabetics

A real world trial featuring an automated insulin pump yields success for those living with Type 2 Diabetes
09 August 2021

Interview with 

Charlotte Boughton, University of Cambridge


2021 marks 100 years since the discovery of insulin. This week, there’s exciting news for the millions of people living in the UK with Type 2 Diabetes who already are, or could become, reliant on injected insulin to control their blood sugar levels - a fully automated pump has been trialed in outpatients for the first time to great success. Charlotte Boughton, a leader on the study, from the University of Cambridge is joining us now... 

Adam - So Charlotte, why was this study done? Why is, you know, just manually injecting insulin, not enough. Why do we need the pump?

Charlotte - We've previously shown that a closed-loop system or an automated system can actually improve people's glucose control in those with Type Two diabetes when they are admitted to the hospital. And so in this study, we wanted to see if the same system can be safe and effective compared to people's usual insulin injections in the home setting, as you mentioned. And we did this study in people with Type Two diabetes who also have kidney failure and need dialysis, and that's because that's a time when diabetes can be particularly challenging to manage

Adam - What are people wearing? What does it look like when they put it on? And how does it work?

Charlotte - A closed-loop system comprises three separate parts or bits of kit. So people wear a glucose sensor, which is about the size of a two pound coin, and that's worn on the arm or stomach, and that measures the person's glucose levels in real-time. And that then communicates with an algorithm or a controller, which in our system is on an app on a normal Android smartphone. And the algorithm calculates the right amount of insulin and tells a small insulin pump, which is a light, slightly smaller than a mobile phone-sized pump that delivers the insulin to the person. And it repeats this cycle about every 10 minutes, constantly adjusting the insulin dose in response to the glucose levels. And the person doesn't need to do anything, they just need to be wearing the kit.

Adam - And how easily do people tolerate this? Because I imagine at least part of it is going to have to be, you know, at least under the skin

Charlotte - Young children who have Type One diabetes wear very similar technology to this, so it's been designed in a way that it's highly acceptable. So the sensor is inserted every 10 days and it's just a tiny little wire that sits under the skin. It doesn't hurt at all. And then that sticks on with a bit of adhesive. And similar to the pump people are used to doing insulin injections, where with a pump, you just need to change the needle every three days, rather than doing injections several times a day. So people really like wearing the device. It takes some getting used to but overall is pretty well tolerated.

Adam - And is that then the big advantage that it takes the thinking, I suppose, out of your head? You can't forget. You can't make a mistake.

Charlotte - Yeah, absolutely. So in this system, it's a fully automated system. For people with Type One diabetes, they use a slightly different system. But in this case, you can almost forget about your diabetes because you don't really need to do anything. You don't need to measure your glucose levels because that's done automatically and you don't need to calculate the right dose of insulin to give, and provided that you change over the devices every sort of few days, then actually the system will work and allow you to get on with your day to day activities.

Adam - Now, to me, that sounds brilliant. The thing I can forget about until something, say, goes wrong. So what if my phone dies or I were to forget it and I needed my insulin. What would happen then?

Charlotte - We rarely leave our phones places, but absolutely it's something that we have to think about when designing these systems. So if the phone's out of Bluetooth communication or the battery runs out, then the pump still gives insulin. It's just it won't give it in an automatic way. So you lose the brains behind the system, but you can still measure the glucose levels and the insulin is still given. So it's sort of a safety net really.

Adam - And it will be effective enough until they can get their phone, or recharge?

Charlotte - Exactly, until it's recharged, yeah.

Adam - That's why it's potentially great for the patient, but what about on the other side, the hospital system, the NHS? Are there benefits for them, for people using this system?

Charlotte - So our study that we've just done was quite small and too short to demonstrate any clear benefits to the NHS. So this would need a much larger and longer study to see if that can translate into improved long-term outcomes, which may have cost savings for the NHS in this population.

Adam - Okay, that makes sense. You mentioned earlier that there are previously pumps for Type One diabetes, and this has been developed for Type Two. Why is Type Two coming later? What's different about it that makes it harder to get a grip on?

Charlotte - It was more where the priorities lay. Some of this research started off in children because of the concern about the effect of high and low glucose levels in young children with Type One diabetes, where they can't often communicate that they've got high or low glucose levels. And these closed-loop systems are transforming the management of Type One diabetes. I think it's come later in Type Two diabetes because in Type One, people don't have any insulin, and so rely on insulin completely to manage that glucose. Whereas in Type Two diabetes, we do have other treatments, so diet and exercise and oral medications and other injections as well. So there are more options available.

Adam - What state are you in? Could you see this being rolled out more widely?

Charlotte - So we already do have similar systems being rolled out in Type One diabetes. And our study was a sort of early proof of concept study that this is potentially a treatment that could be rolled out more widely, particularly in a follow-up study that we're doing in people with Type Two diabetes who don't have kidney failure. And that will hopefully provide evidence to support wider adoption of this technology.


Add a comment