ADA 2017: Type 1 highlights (part 2) – Emily Burns
This is part two of three-part series on Type 1 research highlights from the American Diabetes Association’s 77th Scientific Sessions.
This time, we’re covering technology, from pumps to mobile health.
(Note that some of the research mentioned here hasn’t yet been published and reviewed by the research community).
Unexpected news: pumps and pregnancy
Dr Denice Feig opened the conference with unexpected results from the CONCEPTT study. CONCEPTT is looking at the benefits of continuous glucose monitoring (CGM) in pregnant women with Type 1. While the actual results aren’t ready yet, Dr Feig talked about a sub-study, looking at insulin pumps vs multiple daily injections (MDI).
The idea is that a pump can prevent the blood glucose highs and lows experienced with MDI, keeping levels within ‘target range’ for longer.
The CONCEPTT sub-study set out to find out if using pumps during pregnancy improved HbA1c and the health of the baby, in comparison to MDI. The data is unpublished so we haven’t got much detail, but the bottom line is that they didn’t. The team actually found that there was a larger reduction in HbA1c in the MDI group by the end of the trial.
That was unexpected! Dr Feig suggests that it might be due to MDI users changing settings and injection sites more regularly. She also noted that it’s an observational study, meaning we can’t draw any direct links, and perhaps a clinical trial is needed to get the final answer.
Either way, we’re looking forward to seeing the full results of the CONCEPTT trial.
Predicting your lows
This time last year at the ADA, the artificial pancreas – or closed-loop system – was huge news. This year, researchers are moving beyond feasibility into innovation.
Dr Gregory Forlenza took us through the Predictive Low Glucose Suspension algorithm, or PLGS. Current closed-loop systems often rely on a threshold suspension: the pump stops releasing insulin when blood glucose levels fall too low. The problem with this is that people still experience hypos – we need insulin release to stop earlier. This is where PLGS comes in: insulin delivery stops before levels reach a certain point, as the algorithm predicts the direction blood glucose levels are moving.
They’ve found that the algorithm (used with a t:slim pump and Dexcom G4) could reduce hypos by 50 percent without increasing ketoacidosis.
Dr Peter Jacobs went one step further, with new data on a dual closed-loop system: a pump that releases both insulin and glucagon. Their pump was designed to help people avoid exercise-related hypos – something many people with Type 1 diabetes worry about. The device was made up of a heart rate and activity monitor, a t:slim pump and Dexcom CGM. They looked at the impact of the system on the time spent in hypoglycaemia or ‘target range’ after exercise.
They found that the participants using the dual system spend less time in hypoglycaemia (around 1 percent of the time), compared to the PLGS system (7.3 percent) or the standard system (6.25 percent).
Dr Jacobs believes this could be particularly useful for people who are unaware of hypos or really active. Different forms of exercise have different effects on glucose levels, so we need machine learning techniques to be developed so that the type of exercise can be automatically detected.
Current systems use insulin because of the cost of glucagon and its limited ability to dissolve into a solution. This means that trial participants had to make up new glucagon solutions every 12 hours. For this to become a feasible option, we need a stable glucagon formulation.
Getting closer to closed-loop for children
Finally, we heard about a new closed-loop system for children (aged 6-12). Professor Bruce Buckingham took us through the results of his latest trial to test the new Omnipod system: a combination of the Omnipod pump, a Dexcom G4 sensor and a personalised algorithm.
Twelve children used the system during a 36-hour hospital stay and found that it was safe during the day and night. Now we need further studies to evaluate the system over longer periods of time and in real-life situations.
All on the same CGM page
CGM is becoming more widely used, but each software takes and reports measurements in different ways. This means that neither healthcare professionals nor people with diabetes can get the full benefits.
A team of experts across the world – including people with diabetes – got together to propose standard measurements for everyone. This isn’t the first time there’s been a call to standardise CGM measurements. In fact, there have already been nine consensus statements involving 100 experts, all either covering or suggesting different things.
As one of the team members, Dr Richard Bergenstal, puts it, “We probably don’t need any more consensus meetings; it’s time to align.”
The team identified 14 measurements they believe need to be standardised, from defining what “time in range” really means to establishing how much data is needed to inform clinical care. And it’s not just the measurements, reporting styles must be standardised as well, so they can be understood by patients and healthcare professionals.
What’s App with mobile health?
There are thousands of apps now available to help people manage and treat a whole range of conditions. And with so many people owning a smart phone, mobile health is no longer a thing of the future – it’s here.
And why are mobile technologies better? Dr Shelagh Mulvaney and designer Sara Krugman argued that they provide access to infinite amounts of data, we can study the world someone actually lives in, and we’re much closer to a person’s everyday life. This means we can make more accurate observations.
And so much data is now available to us: location, movements, blood glucose levels, insulin levels, meals. Developers are searching for ways to create even more discreet methods for collecting data without the wearer having to do much.
But lots of questions that need to be asked when developing an app often get overlooked. It’s important that developers and researchers get out there and speak to those who might use it. They may have a great solution to a problem, but according to people affected, they haven’t understood the problem and their solution isn’t right.
But if apps are the way forward, why do thousands exist that no one is using?
Dr Emily Seto reflected on her 10 years’ of research in mobile health (mHealth) to give us some advice. “mHealth technologies aren’t drugs,” she started, “That might seem obvious, but people often treat them like they are.”
Firstly, implementing something in real-life is the difference between success and failure. Running clinical trials is completely different to a real-world environment. Secondly, not all mHealth tools are equal. According to Dr Seto, there are over 165,000 health apps out there, with over 1,100 for diabetes alone. But just because two apps offer the same benefits doesn’t mean you’ll see the same effects. Each app has different features and works differently. She argues that while large reviews have suggested that certain mHealth strategies have fallen flat, they’re not comparing apples with apples.
Evaluating mHealth tech is also really different to drugs. It can take around 17 years to get from early drug research to an accessible treatment. Tech moves much faster: 17 years is the difference between an iPhone 8 and a Nokia 3210! This means that apps can ultimately become obsolete by the time evaluation is finished.
But there are solutions. Firstly, apps can have far more users than would ever be possible in a clinical trial, sometimes into the tens of thousands. While analysing that amount of data has its challenges, it also opens up possibilities for doing large-scale evaluation with analysis happening in real-time.
And last but not least, people don’t get addicted to mHealth apps. You might argue that people are addicted to their smart phones, but Dr Seto points out that they’re usually not sat there on a health app. People often stop using them after the ‘honeymoon’ phase has worn off, which means developers have to make the app valuable – or use incentives. Things like gamification, rewards or social networking can make apps more appealing to a younger audience.
And when asked whether mHealth is a friend or bystander in diabetes management, Dr Seto believes that – if done in the right way – it can be a BFF.