Hello everyone and welcome to this 81st session of the American diabetes Association's annual meeting and to this symposium on translating trial based evidence and real world data for sensor based glucose monitoring to optimize health metrics in persons across the diabetes spectrum. This is a CMB certified symposium jointly sponsored by the University of Massachusetts Medical School and CMI Education Resources with commercial support from an educational grant from Abbott diabetes Care. I'm dr James Gavin and I'm pleased to join you today as program chair and I'm joining a distinguished faculty consisting of Dr Earl Hirsch from the University of Washington Medical Center in Seattle, Doctor Eden Miller from the ST Charles Hospital in bend Oregon and Professor Vivian fonseca of the Tulane University Health Sciences Center in New Orleans Louisiana. And these accomplished speakers will be providing you cutting edge information that is clinically relevant and useful. This entire program and additional CMI offerings will be available as a Clinical Excellence webcast on multiple clinical websites that include www dot diabetes cast dot com. The PowerPoint slides from today's program will be available for your download and any clinical questions about diabetes management may be submitted to W W W dot i Q and a dot a dash c m e dot com. I'm going to start our program today by talking at a high level on the intersection of trial data and real world evidence for the application of sensor based glucose monitoring technology for the diabetes as well as the primary care specialists. Here are my disclosures for your consideration. All all of the faculty members will be providing their disclosures for your consideration today. Let's begin by reminding ourselves of why there is such urgency for improved diabetes management. The stakes are very high and the consequences of poor management are broad based, their debilitating and they're quite expensive and they cut across the gamut of hard related kidney related outcomes and they even um have effects on the longevity issues. Inpatient patients affected by diabetes. And we're reminded while glucose control for prevention and attenuation of complication is not sufficient but clearly necessary in diabetes management. And we see from data like this composite slide from many studies that illustrate what happens when you you see people who receive intensive glucose control. They get to go early represented by the dotted blue line on this slide compared to persons who have a delay in the intensification of their treatment. The difference between those two groups is represented by the shaded area which is really the legacy of disk lissy mia. And it's this legacy of this criteria that drives the consequences that you see highlighted with increased M. I. Strokes and other cardiovascular events. So what it means is that we are trying to manage multiple risk factors in diabetes. Um and it really is function of how well we control the risk of hypoglycemia at the end of the day, balancing that against effective and stable glycemic control. That really serves as a benchmark for how we are managing the overall risk factors that are important in diabetes. Now to better illustrating the intersection of all of the important issues that surround appropriate monitoring data and information that we need for optimal management of diabetes. Let's look at a patient. Uh and I'm going to introduce you to John and we'll come back to John. Over the course of uh this presentation. He's a 58 year old guy, lives with his wife non smoker. He's obese has had diabetes for about 10 years but he's had erratic self monitoring of his blood glucose. He was started on cellphone Neil Yuria six months ago. He is now on Metformin and self o'neill Gloria. And he's been having symptoms of hypoglycemia when he misses meal or with exercise. His wife became very concerned one morning when he seemed very confused and he was sweating profusely. He was advised by his primary care providers to consult with this local endocrinologist to figure things out about this hypoglycemia. One of the things that this highlights is the degree to which there is interaction required between providers to make sure that we get to the real root causes of problematic situations in patients like john Now johN has experienced symptomatic hypoglycemia, which is one of the more disruptive outcomes in the overall management process of diabetes because it can interfere with patient confidence, it can impair treatment adherence. And in fact, uh these are precisely the patients who are most likely to reduce or discontinue their diabetes medications. Now, the fundamental challenge for improving glucose control in diabetes uh and circumventing the complications that are related to hypoglycemia and sustained hyperglycemia is our ability to reduce and control glycemic variability. Now we have now developed metrics for both the description and quantification of the elements of variability. And you see them pictured here on the X axis, we see time and we have now metrics that include time in range time at high levels of glucose, time spent in the normal range, time spent in uh low glucose ranges and on the Y axis we have the variability metrics of things like coefficient of variation or and standard deviation, all of these issues that really reflect how high and how low the variability extremes are. Now, it turns out that the evolution of new technologies and new treatments in diabetes really provide the opportunity to solve persistent problems, one of which of course, is glucose variability. Um and this is especially an issue when it contributes to hypoglycemia. one of the key problems as we see illustrated in our patient john now in order for providers across the spectrum, whether their primary care providers or specialists to know what's the optimal treatment intervention for an individual, we need more comprehensive, quantify quantified glucose information. New technologies like C. G. M. Can help meet these needs when they are properly used. Now, the guidance that we get and the insights that we get about using these new tools usually derives from randomized controlled trials are CTS. Now in every case avoidance of hypoglycemia is going to be a key goal in glucose management. But what is required to achieve this kind of goal in the setting of our CTS may not adequately reflect what is necessary in real world circumstances. So what we've seen then our advances in C. G. M. And this has really radically changed our capacity for improved management of diabetes. Through the use of tools like the ambulatory glucose profile about which you're going to be hearing a great deal more in subsequent uh presentations and this symposium. Um we can actually now allow the detection of if there's a problem and where is the problem and what's the extent of that problem? For example, if you look at the data from the hippo. D. Study at two different groups, the controls and those who use who used C. G. M. There was a 72% reduction in hypoglycemic events in those persons who use C. G. M. Compared to traditional self monitoring of blood glucose. This is a phenomenal outcome and and hypoglycemia is not a trivial clinical concern and that's reflected by data like these from accord. Where you see that there's almost a threefold difference, an increase of three fold in mortality rate in those persons who experienced severe hypoglycemia. Now the benefit of uh glucose uh sensor based glucose monitoring in this case, freestyle or sensor based flash monitoring is has been demonstrated across the spectrum of diabetes type one and type two. We see our cts like impact and replace where uh in these randomized trials we see evidence as depicted here uh where the change in time spent in hypoglycemia markedly reduced in both impact for type ones and in replace for type two. So the interventions with C. G. M result in clinically meaningful and highly statistically significant reductions of time spent in hypoglycemia compared to the controls. Now, one of our great challenges has been how to explain why there's often a difference in the quality and scope of outcomes from an intervention uh as it was used in an RCT compared to its use in the real world. Now, what we now can do is uh design real world trials by uh comparing a trial population and the real world population using propensity score matching. So you're comparing apples to apples and doing appropriately linear regressions. And in that way you can determine what are the contributors to the efficacy gap. That can often be seen. What we see in clinical trials shown here in blue is that both of these agents show efficacy with the GLP, one showing more robust clinical efficacy in terms of a one C reduction. But when we use these same interventions in real world settings, now we see that there is a tremendous gap in terms of a reduction of their apparent efficacy. Our challenge is to figure out why is that? So why is there this gap? And it turns out that when we do a deep dive to figure out what is it that accounts for that gap? The predominant influencer of that gap is poor adherence, uh baseline characteristics and other drug therapies contribute contribute a small amount to that. Yeah but mostly it's at here it's there are so many different factors in the way these new tools and technologies are used in our Cts vs. Real world use and that includes um access motivation for they used the persistence, the adherence, the quality of the data capture the heterogeneity of populations. Generally, healthcare providers have their expectations about a particular intervention shaped by the results of the R. C. T. S. That's the way the interventions are introduced and this is across the spectrum of health care providers, specialists or primary care providers. Real world experiences make off a sharp refinement of these goals because they give us a better sense of what's likely to be the clinical effectiveness. And so real world evidence may provide important and otherwise unavailable data on the effectiveness of an intervention for individual patients. Um and give us some uh insight on what are the necessary capabilities for patients in order to optimize therapy. Uh and that can be uh information that's useful across the full spectrum of our patients with diabetes. So let's get back to john john followed through he discuss his problems with the endocrinologist. It was decided to start him on C. G. M. Using Flash C. G. M, which was covered by his insurance. He's been on it for two months, has not experienced another hypoglycemia episode. He's learned to use trend arrows and he does frequent scanning. And and this is encouraged by both his PCP and his indo. Now this uh means that the higher frequency scanning that is available by using something like flash glucose monitoring is something that has particular impact. And it has high impact for vulnerable patients like johN. And that doesn't start with type two patients is for type one patients, it's for any patients in whom reduction of hypoglycemic risk is of central importance for people who having sick days. Um For elderly frail patients, we can encourage them to scan offer frequently especially when uh they are in situations where they are at the greatest risk for hypoglycemia and particularly when they're about to do things like increased physical activity. It turns out that the evidence that the tools of C. G. M. Can make a big difference in hypoglycemia risk is quite compelling. Here. We see data from real world data From the first sensor use of almost 15,000 persons. And what we see is that there was a significant reduction In hypoglycemia across the entire spectrum of severity of hypoglycemia. In fact, at the most severe levels of hypoglycemia, there was a 74% reduction observed in the first two days of sensor use. It doesn't stop there. Remember our big challenge in patients across the spectrum, Type one, Type two for all of us as providers, whether we're primary care or endocrinologists, we want to mitigate variability. So it's not just the lows that we are concerned about. We want to reduce hyperglycemia as well. And this has been a particularly uh important metric during the covid pandemic. Where the presence of persistent hyperglycemia had such a severe, deleterious effect in persons with diabetes who were made more susceptible to covid and who had worse outcomes. It turns out that after a period of weeks using um see Gm. In this case, flash glucose monitoring. We see that there was a substantial and significant reduction in hyperglycemia as well. And so one thing that many of these studies have shown us whether we're talking about R. C. T. S, but even more so real world studies is that in the case of sensor based C. G. M more is better. You should encourage high scanning frequency to optimize diabetes management. Now, the information that we have gotten from glucose monitoring methods really strongly affirmed that um what we really need is the kind of information that can only really be gotten optimally from C. G. M. Because a one C while important is the standard of care across clinical disciplines. It does not reflect daily highs or loads. Uh You cannot act on what you don't measure the fluctuations throughout the day or not known. Um self monitoring of blood glucose, the same kind of limitation. Uh You only get information before the times when you're testing. If you miss an important point of hypoglycemia or hyperglycemia, you can't act on that information because you didn't get that information. C. G. M. Now facilitates digital connectivity data transfer across the spectrum of providers. We can now take advantage of large databases like electronic health records. It takes time. It could be time consuming. But one of the things that we find is that each form of C. G. M. We need more real world data and and there are efforts that are underway um to to generate more real world evidence to improve access and two more um comprehensively assess impact. This is an illustration of what's happening. For example, in the United Kingdom, they have a county based surveillance system in place where they are now surveying Type one patients to see who has access to uh C. G. M. But the structure that they put in place for this surveillance system is a structure that can actually be used to generate data on not only who has access but who's using it, What kinds of data is being generated and in whom do you see the evidence of meaningful benefit? So we're seeing significant progress being made. Now the metrics that have been developed and standardized of course are highlighted by the ambulatory glucose profile and you're going to hear a great deal about this particular metric and how it is used in clinical assessment later on, particularly from Professor Fonseca and what this does is that it sets up what the target range ought to be. This is what we should be aspiring to. The low and very low regions. And the high and very high ranges are the areas we're trying to avoid, we're trying to minimize exposure there. The greatest urgency is frequently. Um First and foremost to prioritize those areas of low blood glucose, The hypoglycemia have the greatest urgency. Um And then of course we have to make sure we pay some attention to how much time. And what is the extent of the hyperglycemia? Excursions as well. So these metrics give us all the things that we really need to know in order to define whether there's a problem. How severe the problem is, how long the problem last and what is it that urgently needs our clinical attention? So time in range targets become a centerpiece of continuous glucose monitoring systems and a summation of what it is that we'd like to try to capture with ambulatory glucose profiles is represented here. You can look upon this as a desire to move from black to green. If you look at the elements of an ambulatory glucose profile that we desire, we wanted to be green. So here the profile is not flat, it's not narrow, it's not in range. So nothing about this profile really lands on what we want. This next profile shows that it's certainly flattered and we wanted to be flat. That's a desirable characteristic, but it's not narrow and it's not, here's one that's not only flat, but it's also narrow, but it's not in range. Ultimately, what we want is more green. You want it to be flat, narrow and in range because that is the way Mother Nature has designed our glucose profile under non diabetes conditions. So if we go back for a second to john, he's a little older now. He's got hypertension. Just left academia in addition to his diabetes, you're still frustrated with the challenges of taking multiple medications trying to keep everything under control. But he is most encouraged by how he has managed his diabetes. Especially with his use of C. G. M. He's beginning to have some additional issues but he is clearly confident that he will keep things in control. Okay now it has to be noted that the full impact of the benefits from C. G. M. Are not readily available too many or even most of the patients who really need it because of restrictive utilization policies. There are policies in place that are barriers to access that include the requirement that the patients be treated by an in endocrinologist. Um They are um injection thresholds, they have to be getting a certain number of injections a day. Have to have things like hypoglycemia unawareness or nocturnal hypoglycemia. There are many of these restrictive barriers that stand in the way and they are not only scientifically uh not credible. Uh they are often uh poorly documented policies. So what we're trying to do is to get around those kinds of things so that we can leverage the benefits of the evolution of of glucose monitoring. Things like flash glucose monitoring represents a way to get the dense data that we need for complete interpret herbal glycemic pictures across the spectrum of diabetes. Mhm. Its uniqueness is in part because of the painless frequent scanning that can be done. There is no user calibration required, no routine fingerprints and it automatically measures, captures and stores the data. Digital connectivity is widely available and is easily used by all health care providers. Whether primary care specialists, all of us now will have the benefit of this tool. And what we've seen with real world evidence is strong complementary information that allows identification of the most appropriate patient targets and the optimal clinical settings for the use of these agents. So what we can say, finally, is that the guidelines that have been developed have been developed to help us solve problems, prevent complications and to optimize the quality of life in every patient. But it's very clear that the the achievement of the goals of care that we are interested in depend on adequacy of appropriate testing. It is essential uh that we have the information that we need and it is also essential that provide us from different specialties as happened in the case of our patient, john uh collaborate to optimize patient care, taking full advantage of all of the information that is now available through sensor based glucose monitoring. So with that as a high level overview, I'm now going to say thank you very much for your attention, and I'm going to now turn the stage over to my colleague, Professor Earl Hirsch, for the next presentation in this symposium. Thank you very much.
Related Presenters
