So Chair, ladies and gentlemen, my name is Emma Wilmott and I'm an associate professor at the University of Nottingham in the UK. I'm also a big technology enthusiast. I'm the founder of the ABCD Diabetes Technology Network UK and I'm delighted to be here today to speak to CGM transforming outcomes in people with non-intensive type 2 therapy. These are my disclosures. So, the next 25 minutes or so, I'm going to walk you through the evidence base for continuous glucose monitoring, considering it in insulin treated people and also non-insulin treated people with type 2 diabetes. Now I know that just sticking with the evidence can be a bit arduous, so to make sure you stay awake, I shall also throw in some interactive case studies. And then I'll finish by considering emerging evidence and where we go next. I want to start by reflecting on the amazing journey that we've been in the care of people living with type one diabetes. When I first started as a consultant in diabetes, I spent a lot of time in my clinics encouraging people to do more blood glucose monitoring and often it'd be a fruitless conversation and we'd struggle to get the data that we really needed to support optimal outcomes. Myself and Lala Lee Rathna got funding from Diabetes UK to deliver the FLSH UK randomized control trial to evaluate flash glucose monitoring in people with type 1 diabetes. We were able to demonstrate significant improvements in HbA1c, but also less hypoglycemia and improved treatment satisfaction. But you and I both know that often the evidence base alone is not enough to secure reimbursement. And a crucial part of this clinical trial was the inclusion of health economic information. And that evaluation showed that flash glucose monitoring was highly cost effective, with a cost per quality of 4,477. But crucially, for people going onto flash glucose monitoring with a very high HbA1c, the device was shown to be cost saving. And the wealth of evidence from type 1 diabetes has really transformed care where I work in England in the United Kingdom, where we now have more than 95% of our type 1 population having access to continuous glucose monitoring and leading to improved outcomes across the board. But it's also fundamentally changed the conversations that we're having in clinic. Gone are the days of looking at blood glucose diaries and HbA1c. Instead, we're now focusing in on really thinking about what the time and range is. And finally we have that objective measure of time below range in hypoglycemia. It's a very different place in the world of type one diabetes. Now sadly, the same cannot be said for type 2 diabetes, while we've had amazing advances in the therapies available, certainly in the UK where I work, access to continuous glucose monitoring remains fairly limited. In the UK to get access to continuous glucose monitoring, people need to have problematic hypoglycemia, whether that's severe hypos, impaired awareness. They need to have an issue with being able to check their blood glucose, or they need to be somebody that's checking 8 times per day, so fairly limited uptake overall. And you'll be aware that there's been a wealth of randomized controlled trials of continuous glucose monitoring in diabetes, and I thought I'd recap on some of the key trials to be aware of, and I'll start with those of intensive insulin treatment. So Roy Beck and colleagues way back in 2017 published the Diamond type 2 study, which took people in America on multiple daily injections with type 2 diabetes with a high HbA1c and randomized them to CGM or standard care. And you can see on the right-hand side here that the between group difference in HbA1c was 0.3% in favor of continuous glucose monitoring. Around the same time, we also had the Replace BG study here demonstrating the impact of flash glucose monitoring in those with intensive insulin therapy across Europe. And although this didn't meet the primary outcome of a difference in HbA1c, we can see at the top right that when you look at those under the age of 65, there was a between group difference in HbA1C in favor of intermittently scanned CGM. But for me, the fundamental take home from this study was a dramatic improvement in time below range at all the various thresholds as shown here on the right-hand side. Another debate which rages is when we give people CGM, do we need to be providing simultaneous education? And this study by Kim, which was published in Dit Babetolosia just a couple of years ago, I think really answers that question for us. They took people with type 2 on multiple daily injections and randomized them to either continuous glucose monitoring with structured education, specifically thinking about how we support people to dose adjust their insulin and how they think about the timing of. Their insulin when looking at CGM patterns and compared that to CGM alone and also self-monitoring of blood glucose. And as you can see on the right hand side here, that structured education around the specific dose and timing of insulin was crucial in terms of delivering additional benefits. So I would argue that yes, education with CGM definitely does matter. So what about the next category, what about people with type 2 diabetes that are on just basal insulin alone? Well, of course, the landmark mark study looking at CGM in the context of those on basal insulin only was of course Thomas Martin's mobile study. And here they took people with an elevated HbA1c from 22 practices in the USA and randomized them to CGM or control. And again, we see that between group difference and HbA1c this time at 0.4% in favor of CGM. And you'll see at the bottom right-hand side here that there was a corresponding improvement in time and range and also a significant improvement in time below range or hypoglycemia. But a really intriguing aspect of this was that they saw these benefits with no between group difference in total daily insulin dose, which raises a question about whether this is actually behavioral changes that are really driving the significant improvements in glycemia with access to CGM. And more recently, just last year, they've published their three month data from the mobile study, which I think really helps us to start to unpick and understand what's happening when people get access to CGM. And on the left-hand side here, you'll see that those in the CGM group improved their HbA1c from 9.1 to 8% compared to 9 to 8.5 in the blood glucose monitoring group, with a whopping 0.6% between group difference in HbA1c. But when you look at what's happening to the time and range over that three month period, you'll notice that from baseline into week one, time and range improves from 41 to 50%. By week two, it's increased again to 56%. And this is happening during a time when they're not seeing the healthcare professionals, again, lending to support to that argument that this is participant driven lifestyle change that's possibly driving those improvements in glycemia. And then another study which I always like to highlight, it really fascinates me, is Coe's PDF study. This is a 12-week randomized control trial of people on oral and basal insulin therapy with an Hb1c above 7%. They were randomized to CGM or control and followed up for a period of 3 months. And you can see by 3 months there's a 0.5% between group difference in HbA1c, again in favor of the CGM group. But what I particularly love about this study is how they achieved this. They developed a really, really simple tool to support participants in the study. They were asked to consider two key things. One, is the food you're eating healthy? And two, what does it do to your post-meal glucose? If it's a healthy food, and it doesn't spike your glucose after a meal, fantastic, stay on current eating pattern. If it's not a healthy food, but it doesn't spike your glucose, just cut down on the amount a bit. If it's a healthy food, but it does spike your glucose, cut down on the amount. And actually if it's unhealthy and it's spiking your glucose, you're best to stay away from it. So really simple er advice that I certainly use in my clinical practice today. So, if you've had enough of listening to the evidence and you want to be taken back into the clinic, here is a clinical case, and I'd like you to have a look and consider what your treatment strategy would be. So this is a 4, 52 year old female with type 2 diabetes, she's a smoker. She's got a BMI of 29, and her HbA1c is up at 9.6, and it's been there for some time. She typically only checks her blood glucose for a few days before she comes to see you in clinic, and it's typically first thing in the morning. Her current treatment for her diabetes is her Humulin I insulin, 19 units twice a day and Eagliflozin. You've tried various things in the past, she's declined a statin, she's been intolerant of metformin, intolerant of GLP-1, so this is where you're at currently. So, here she comes to clinic and these are the glucose data that she brings with her, so we can see the glucose levels are generally in target there, 4.2, 5.2, 6.3, 5.1. But this is in the context of a very high HbA1c, so what are you going to do? How will you alter the insulin dosing? Will you 1, increase the morning dose, 2, increase the evening dose, 3, add in prandial insulin, 4, wait for more blood glucose data first, or 5, wait and give a trial of CGM first. Well, Here we have her CGM data, and what this reveals is that actually she is indeed reporting correctly on her blood glucose levels in the morning, which are indeed tending to be on target, but what it does show is what we're missing. Look at the elevations in the glucose levels across the day, the postprandial hyperglycemia, that the occasional blood glucose checks purely are not picking up on. And actually in this scenario, this lady came into our clinical trial and had access to continuous glucose monitoring. And with that information from the continuous glucose monitoring, she was empowered to make substantial changes to her diet. She told me that she gained insight into the impact of foods on her glucose levels and she learned to avoid those foods that spiked her glucose levels the most. She came back saying that she'd cut out sweets from her diet, she was avoiding chocolate, and she was reducing her portion sizes for things like bread. She had actually had to reduce her insulin dose as a result of the dietary changes that she's made, and as you see on the right hand side here, her time and range improved from 41% up to 85%, with a corresponding improvement in HbA1c from 9.6 down to 6.9%. But I can hear you thinking, well that's great for 4 months, but is this sustainable? Um, and we can have a look at what happened in the longer term. But fundamentally you can see from the traces before and after what she was able to achieve by addressing diet, reducing those postprandial spikes, improving time and range up to 85%, and improving her time and range to 6.9%. After 8 months, she was indeed able to sustain those changes she's made. By 8 months she'd had to further reduce her insulin doses, she had made sustained improvements in her dietary intake and was maintaining 85% time in. Range and you can see here that her Hb1c improved immediately and was sustained all the way up to 12 months with a reduction in weight from 96 to 87 kg, with no additional therapies for her type 2 diabetes. So definitely a success story and not obviously not representative of everybody we see in clinic, but it certainly shows us what is possible when we give people access to their glucose data and they feel empowered to make changes as a result of that. So moving on to think about access to technology in this very group. So I'm interested in understanding access to technology across the wider cohort. So have you got access to funding for all your patients who are non-intensive insulin therapy, that's basal insulin. Some of you will do, some of you will do, but only if you have insurance and certainly where I work in England and the UK er we do not generally have access to the majority of people um in this category. So where are we in terms of um access for non-insulin therapy and the evidence base supporting that? Well, WADA and colleagues way back in 2020 looked at a randomized controlled trial of CGM in people with non-insulin treated type two. And they were able to provide devices for the 1st 12 weeks, and then both arms returned to standard care up to 24 weeks. And although there was no difference at 12 weeks, by 24 weeks, you see there was a clear between group difference between the two arms at that point, with improvements in treatment satisfaction with access to continuous glucose monitoring. But more recently, you'll be aware of the immediate study, this is a multi-centre randomized control trial of people on non-insulin therapy with an elevated HbA1c. It had a primary outcome this time of time and range, and they followed people up for 16 weeks and explored the impact of CGM and education versus education alone. And what you can see here is that between group difference in timing range of 10%, so, um, both statistically and clinically significant difference in time and range, with a corresponding between group difference in HbA1c of 0.3% and higher glucose monitoring satisfaction in the intervention arm. But as I've hinted to earlier, again, there's this whole question about what's driving this. Is it behavioral change, is it therapeutic change? And again, in this trial, non-insulin therapies were added among less than 10% of the participants in each arm. Again, giving a nod to the fact that this is possibly driven predominantly by lifestyle change. So one further study to highlight was this one published just a couple of years ago, I think the COVID pandemic threw us all into the world of remote telemonitoring. And actually this study answers the question, can you combine remote virtual diabetes educator visits with CGM to deliver glycemic outcomes? And the short answer is yes, this trial over six weeks was able to show a 0.65% greater HbA1c reduction by the period of 12 weeks, and they lost more weight and were more satisfied with their treatment. So that combination of CGM with data in the cloud and remote support can be superior to usual care. So thinking about that group with non-insulin therapy, if you've not got access in people with basal insulin, then I suspect you may not have access in people with non-insulin therapy, but just reflecting on, you know, can you access CGM for people with non-insulin therapy? Is it down to insurance or is this just not a goal in the country that you're in an important consideration for all of us? But what I will say that is in summary, we have got clear level and systematic review and meta-analysis consistently telling us the same message. This one published in 2024 showed a 3.4 millimole per mole improvement in HbA1c across 12 randomized controlled trials. But what was really fascinating from my perspective was that the improvements in glycemic outcomes were the same in this trial, whether or not people were using insulin. This has been further backed up by Tan and colleagues umbrella review showing improvements in HbA1c of around 4.4%, improvements in time and range, and reductions in hyperglycemia. And these findings were found to be invariant of pre-existing insulin treatment, study funding, risk of bias, and the authors from this study conclude that we recommend the introduction of CGM into standard care alongside SMBG for people with type 2 diabetes. And you've heard me mention quite a few times that indication that perhaps it's behavioral change that's underpinning this, and I was interested to see that this recent systematic review and meta-analysis has really focused the attention on continuous glucose monitoring and lifestyle choices and nutrition management. And what they were able to show that when you look at those trials, there's a 0.46% greater reduction in HbA1c with access to CGM versus those in the control arm, and also a reduction in body weight. So high level consistent evidence coming out of the same message that there are improvements in HbA1C in favor of continuous glucose monitoring. So that's where we're at, but a key question is what comes next. So I am the chief investigator of the Freedom to randomized control trial alongside my er co-chief investigator, Doctor Lala Lee Rathna. This study will be presented this week at the ATTD and I'm going to currently just give you a quick overview of the background to the study and some of the baseline results, and I'll invite you to join us tomorrow morning, on Thursday morning to hear the highlights and the outcomes of 4 and 8 months from the Freedom to study. So you'll be aware that an increasing proportion of the people that you support who live with type 2 diabetes are treated with modern therapies such as SGLT-2, GLP-1, and often combination with basal insulin. But you'll be equally aware that despite this combination of therapy, many still do not achieve their glycemic targets. But also, I've walked you through the evidence base for continuous glucose monitoring, but things have moved so rapidly in the world of type 2 diabetes that few of these trials have been done in the context of the modern therapies that we're using today. And also I've given the nod to the behavioral changes underpinning the benefits of CGM, but actually our understanding remains fairly limited. So the aim of the Freedom 2 trial is to determine whether the Freestyle Libre 3 CGM system improves HbA1c over 16 weeks compared to self-monitoring of blood glucose in adults with suboptimal glycemia. This is a UK multi-center trial across 24 sites, and we recruited people with type 2 on basal load-only insulin, and either SGLT-2 or GLP-1 with an elevated HbA1c. Patients were recruited and during the 1st 4 months of the study, they were encouraged to titrate their insulin and make lifestyle changes aiming to get to glycemic targets. If by 16 weeks they had not achieved those targets, they were then supported by healthcare professionals to consider the addition of further agents to support achievement of the optimal outcomes. We also had an extension phase for those in the control arm who got access to 4 months' use of the FreeStyle Libre at the end of the study. So the primary endpoint of the Freedom to randomized controlled trial is the between group difference in HbA1c at 16 weeks. A key secondary is HbA1c at 32 weeks, and we're looking at a range of other outcomes such as objective accelerometer data about physical activity, CGM metrics, changes to therapy safety, and of course patient reported outcomes, and we had Prof Kaf Barnard doing qualitative interviews as well. This was a huge trial of CGM in people with type 2 diabetes. We enrolled 469 participants, of which 303 were randomized, 198 in the intervention arm, and 105 in the usual care arm. Here I present the baseline data from Freedom 2. You'll see that the average age in the trial is 61 years and 2/3 were male. The average BMI was in the obese range, and there was a long duration of diabetes, which is reflected in the prevalence of complications as shown here on the right-hand side. At baseline, our two groups were very well matched for HbA1c with an HbA1c of 8.8% in both arms, and they were doing on average around two finger stick glucose tests per day. You'll see here that at baseline, the time and range was 40%, and very low rates of hypoglycemia, 0.5% time spent time below range. So, that's a little teaser on the baseline data. Um, I will be sharing the 4 and 8 month data very shortly. Ultimately, the Freedom 2 trial is going to provide us with much needed information on the efficacy of the FreeStyle Libre system on glycemic outcomes in people with type 2 diabetes and suboptimal HbA1c levels, despite modern day therapies. Our two-phase design will provide insight into the impact of patient-driven insulin initiation and titration and therapy escalation, and also the accelerometer and diet information will give us some information about some of the key behavioral changes underpinning that. So before I finish, I just want to to walk you through some key late breaking abstracts that will be presented during the ATD conference this week. So I'd like to highlight Ramzi Adjan and colleagues who will be presenting this work looking at GMI potentially, Overdiagnosing prediabetes and they're arguing that um it needs to be replaced with an updated GMI aligning more closely with the HbA1c, so please join him in hall 212 on Friday morning for more information about this study. We also have this data from Pi and colleagues looking at predicting insulin reduction following GLP1 receptor uh receptor initiation, and again they'll give you some much needed information about which of our patients are at greatest risk of hypoglycemia and more likely to require insulin reduction. Looking at access to CGM and those on basal insulin, we also have this budget impact analysis of the Freestyle Libre systems from Gentle and colleagues, and they are looking at the expanding reimbursement of Freestyle Libre in those on basal insulin leads to a net cost saving of €444 per patient per year. So if you're involved in making business cases, etc. in your country, this might be of interest to you. And also we have these data from Yale and colleagues looking at a focus group of people using freestyle libre and exploring the impact and self-management and communication and stress levels and lifestyle changes over time, so please look out for that also. So Chair, ladies and gentlemen, I've been delighted to join you today to walk you through the current evidence base. I think you'll agree we've got undisputed benefits of uh CGM in type 1 diabetes. We are seeing continuing emerging evidence supporting the role of CGM in people living with type 2 diabetes, and as many of the studies suggest, I suspect there's a lot going on with lifestyle change when people get access to that real-time CGM data. But of course this week, the freedom to randomized controlled trial results will be presented on Thursday, and I would encourage you to join me where I'll be sharing both the 4 and the 8 month results, so I look forward to seeing you then. Thank you very much.
Related Presenters
