Diabetes developments: a look at the latest technologies, medicines and treatments – by Simon O’Neill

Simon O'NeillIn a regular blog series, Simon O’Neill, Diabetes UK’s Director of Health Intelligence and Professional Liaison, rounds up the latest diabetes news.

This week Simon updates on the latest technologies, medicines and treatments.

Smartphone-compatible ultrasound

Smart phones are being put to more and more innovative uses. As well as the various apps many of us use to help manage diabetes, we’re also using them to display CGM and Flash blood glucose readings and even to take retinal photographs. However, the latest technology to use a smart phone, which has just been approved for use in the US by the FDA, is an ultrasound device.

Ultrasound is used in many medical situations to gain a non-invasive view of what is happening in the body – whether to measure the development of an unborn baby or to assess the health of various organs. Traditionally, ultrasound scanners consist of three separate transducers (devices which convert energy from one form to another – such as a microphone converting sound to a digital recording), connected to large units, in order to deliver a complete and moving image. Ultrasound machines have historically been quite expensive and are also not that mobile, meaning that people normally have to attend an out-patients clinic.

However, a company has now produced the Butterfly iQ system which places the three transducers and more than 10,000 sensors in a single, hand-held scanner. The scanner is used to undertake the ultrasound in the usual way but the image, instead of appearing on a screen on a large device, can now be viewed on an iPhone. Images are also then sent automatically to a cloud storage system, allowing them to be connected to a person’s hospital records.

The device has been approved for use in the US in 13 clinical applications. The company are also now working on developing add on software to help users get the best images and to help doctors interpret what they are seeing in the scans. The hope is that this will enable lest skilled users to still capture high quality images to help in health assessment.

Skin as a glucose sensor?

A couple of stories have been reported this month looking at very innovative ways of measuring blood glucose levels. Both are at early stages of development but have exciting potential. Traditionally our skin has proved a difficult barrier to overcome when trying to measure blood glucose non-invasively. But both of these approaches use the skin as part of the solution to the problem.

The first group from Chicago managed to edit skin cells from a mouse and transform them into a glucose detector which can then be re-grafted on to the animal, so there should be no issue of rejection. They used mouse stem cells which would normally make new skin but, using the gene editing technique CRISPR, they added a gene from E.Coli bacteria which makes a protein that ‘sticks’ to glucose molecules, alongside DNA which makes two fluorescent molecules. The idea is that when glucose sticks to the E.Coli protein it causes the fluorescent molecules to move in a particular way that generates a signal easily seen through a microscope.

The skin cells were then re-grafted back on to the mouse. When they were given a large amount of sugar, the cells rapidly reacted and the rise in blood glucose could be measured. The reaction was apparently as accurate as a blood test, in terms of a change in blood glucose – though it isn’t clear yet whether it can tell exactly how much glucose is in the blood.

The big downside is that the change in luminescence in the skin can only be read under a laser light and through a powerful microscope – so it’s nowhere near ready for use in humans. But the research team have further plans to develop the technology and have ideas on how this could become a practical tool for people with diabetes. The team are also interested in whether similar bio-engineering could also help similar stem cells produce insulin on demand, creating a closed loop system – but that’s even further into the future.

The second approach might well suit our tattoo loving generation. A team from Harvard and MIT have developed a ‘smart ink’ that could change colour with rising blood glucose levels by interacting with interstitial fluid, that’s used in CGM and Flash glucose monitoring systems already.

The team call the approach “Dermal Abyss” and they have developed two types of ink so far which they have tested on pig skin in a lab. One ink measures blood glucose and the other measures levels of sodium. The glucose ink changes shade from green to brown as more glucose is detected. Ideally the colour change would be noticeable enough to see with the naked eye, but the team are also developing an app which could give a more detailed analysis using a scan of the tattoo.

The technology is still at the proof of concept stage and there are many issues to overcome. One of these is that the inks don’t last very long, so would need to be regularly applied. Again, getting an exact measurement of actual blood glucose could also be difficult – and one only has to remember back to urine testing strips, where you visually had to decide how much colour change there had been to get an accurate result.

The interesting thing about both approaches is that the skin itself is a key component of the technology. One big advantage of using skin is that there is no power source required to deliver a result and no ongoing running costs.

Artificial Kidney

End stage kidney failure (ESRF), requiring either routine dialysis or a kidney transplant, is one of the most costly complications of diabetes, not only because of the expense of dialysis but also the huge burden it places on people to attend a hospital clinic at least three times every week. Ideally more people would receive kidney transplants – but with over 5,000 waiting for a kidney transplant in the UK in 2016-17 and only 1,366 deceased and 1,009 living donors, many people will have to wait a very long time. Nearly 300 people died while waiting for a donated kidney last year and over 3,000 new patients join the waiting list each year.

So it’s promising news that after over 20 years of work one research team believes they are close to perfecting an implantable artificial kidney. The ‘bionic’ device uses similar electronic chips that you find in smartphones, alongside special silicon nanopore filters combined with living kidney cells. The whole device is then implanted, connecting it to both the circulatory system, in order to filter the blood and the bladder. To prevent rejection of the device, the silicon filters act as a screen, keeping immune cells physically apart from the kidney cells embedded in a micro-scaffold.

So far the device has been trialed in pigs and shown good results. The team are now about to embark on human testing, implanting the device in the abdominal cavity with the first trials scheduled for 2018. The major concern is blood clotting, so only if the devices have survived a month internally without a reaction will the filtering process be switched on. Once that has been shown to work well, the device will be combined with the kidney cell element and be tested as a full artificial kidney. Final trials should report in 2020. Interestingly over 9,000 people have volunteered for the trials, perhaps giving a clear indication of the drawbacks of current dialysis.

180 day sensor

The latest development from Senseonics is an even longer lasting version of their implantable Eversense blood glucose sensor. The current sensor lasts for a very impressive 90 days – but the new version can now give blood glucose readings for up to 180 days, more than 12 times longer than any other CGM type sensor.

The Eversense XL CGM system has already gained approval in Europe and Senseonics is working with Roche to distribute the device in Germany, Italy, and the Netherlands, probably later this year.

Like the 90-day sensor, the Eversense XL is about 1.5cm by 3mm and is implanted in the upper arm under local anaesthetic through a small incision in the skin. The person then wears a rechargeable transmitter on the skin above the implant which can be removed at any time. This both powers the sensor and transmits real time glucose readings to a smartphone. The transmitter can vibrate if glucose levels are going too low or high and also sends an alarm to the phone. As with most other CGM systems, it still needs to be calibrated with two finger prick blood tests a day.

It isn’t yet clear when the device might be available in the UK, although initial trials were carried out in the UK.

An update on the artificial pancreas

With more and more positive data about outcomes with an automated insulin delivery system – including fewer hypos, more time in target range and lower levels of diabetes distress – it isn’t surprising that so many people are working to get their systems to market – and they aren’t all big corporates either.

Obviously many scientific groups are developing the algorithms needed to make the artificial pancreas work. Diabetes UK have been funding the work of Dr Roman Hovorka and Dr Helen Murphy at Cambridge, but a quick look at ClinicalTrials.gov shows at least 133 ongoing trials into this sort of technology at hospitals and universities all around the world.

So far only one company has launched an early stage version of the AP – the Medtronic hybrid closed loop system, the 670G. This now has limited availability in the US with plans to probably roll out in Europe from 2018, although the device apparently hasn’t yet got a CE mark. Medtronic have also had problems with a global shortage of CGM sensors (an integral part of any AP solution) which may further limit expansion into markets outside the US.

But other companies are also moving forward with their own approaches. Tandem is partnering with Dexcom to turn their t:slim pump into a “hypoglycaemia-hyperglycaemia minimizer” with trials expected to begin in early 2018. Insulet, who produce the OmniPod, are also working with Dexcom to produce the OmniPod Horizon which will be a closed loop system with trials taking place in 2018 and a launch expected in 2019.

Bigfoot Biomedical are working with Abbott, using a modified version of the Freestyle Libre as a CGM sensor, to develop their system with trials planned for 2018 and a potential launch in 2020. Another US Company, Beta Bionics, are also working with Dexcom on the i-Let device, which could be the first dual hormone (insulin and glucagon) artificial pancreas. However it is likely that they will move forward with an insulin only product by 2020. These are all likely to appear first in the US, so may take some time to reach the UK market.

In Europe, Roche are working with Senseonics and their Eversense sensor to develop their system, with trials expected in 2018. The Cellnovo patch pump is also being used with two different algorithms and Dexcom sensors with further European trials ongoing and potential launches in 2018.

Although this is really exciting, the reality of getting these devices into the hands of people with diabetes won’t necessarily be easy going in the UK. Despite current NICE guidance in England for both insulin pumps and CGM, we know that the number of people currently both eligible for this technology and able to access it through the NHS is very limited. As these are the two basic elements of any AP system, the likelihood is that any roll out of this technology will be very slow. The response of CCGs to the announcement that Freestyle Libre would be added to the Drug Tariff is very telling, with many deciding not to prescribe it at all until further evidence is available.

So are there any other approaches to the AP? The simple answer is yes. The difficult bit about developing an AP system is the algorithm. It doesn’t really matter what pump or CGM sensor you use, as long as the sensors are accurate enough and the pump can be instructed to deliver more or less insulin. The algorithm is what does all the work and that is what all those research studies are investigating – what is the best algorithm to instruct the insulin pump to give exactly the necessary insulin to maintain blood glucose at a stable level. For any commercial company, producing a device like the MiniMed 670G, they need robust evidence that their algorithm not only does what it says on the tin but is also safe to use and won’t lead to any patient harm. FDA and EMA wouldn’t approve a product without that safety data.

But what if you’re prepared to take the risk yourself? Welcome to the world of Loop. Loop is an app template which can be used by individuals to develop their own closed loop system. All the software is open source – so available to anyone – and using this you can take many of the pumps and CGM systems you are currently using and create your own AP. The individual takes full responsibility for building and running this system and does so at their own risk. This is all part of the NightScout Movement who are impatient for this sort of technology to be made commercially available so are developing their own resources. The developers are quick to point out that this is an experimental project and is not currently approved by any national body for therapy.

The system can be used as an open loop (where only the basal/background insulin is automated) or closed loop, where the system takes on everything you throw at it, including exercise and meals.

The numbers trying this for themselves is still small – but is growing, particularly as people see the successes that others are having and because nothing else is currently available in most countries. But it is important that more people are aware of it, particularly HCPs who may have some of their Type 1s already using this technology. They might think it is irresponsible or dangerous, but their views won’t stop those who have access to pumps and CGM looking in to whether this is possible or desirable for them

Interestingly JDRF in the US have responded to this open source movement by trying to encourage device manufacturers to make devices that can communicate with one another, in an effort to drive forward the move towards an affordable AP system. They are also trying to tackle the issue of liability, which is why most manufacturers haven’t gone down this avenue before, with the very real fear of being sued if something goes wrong.

i-Port Advance Injection Port

For the many people not using an insulin pump, multiple daily injections are obviously still the norm but this doesn’t mean that they are easy for everyone to do. Although normally not painful, some people are scared of injecting and there have been limited options to help them overcome this phobia or to inject insulin more easily. These have included devices which hide the needle, needle free injectors which blast insulin directly through the skin and auto injectors which make it easier to insert the needle.

However, there is another approach, using the same sort of technology that you would normally associate with an insulin pump. Rather than injecting into the skin each time, the i-Port Advance is applied to the skin in the same way as a pump infusion set and can be worn for 3 days. Any injections are then placed directly into the i-Port, allowing up to 75 multiple injections over the 3 day period. And this isn’t limited to short acting insulin. Long acting insulin can be injected through the same port, leaving an hour long gap between short and long acting insulin – and it can also be used for any other subcutaneous injections, such as heparin or growth hormone.

As well as helping those who have real needle phobia, this may also be a useful device for people with T2D who are starting out on insulin and are not yet very confident in injection technique or for parents of smaller children. There is also the potential to use it in people who have lipohypertrophy (a build-up of fat under the skin in areas of repeated injections). This condition is much more common than previously thought, with more than half of people injecting insulin experiencing it. Lipohypertrophy can lead to more frequent and unexplained hypos and more glucose variability and can be prevented by regular rotation of injection sites, or not injecting in an affected area for several months. Because it reduces the number of injections needed, the i-Port, could help overcome some of these issues.

Unfortunately the device isn’t yet available on prescription and costs about £72 for a month’s supply available online via Healthcare Equipment and Supplies  or through selected Payden’s pharmacies. More information can be found on Medronic’s website

You might also like