Skip to main content

Outage notice:

Due to the global information outage that occurred this morning, Quest is taking swift action to respond, recover and restore our operations. Our Patient Services are operating with reduced capacity so you may experience longer wait and service times. Our Customer Contact teams are currently unavailable. We appreciate your patience as we work toward returning to normal operations. 

Hide

Early Identification of Type 2 Diabetes Risk for the Primary Care Setting

On-Demand Webinar

Early Identification of Type 2 Diabetes Risk for the Primary Care Setting

The culmination of progressive metabolic conditions, including the obesity epidemic, has led to increasing rates of type 2 diabetes. Identifying individuals at risk of progressing to diabetes has become increasingly important. In this webinar, Dr Michael McPhaul will discuss how appropriate laboratory testing will allow for the early identification of metabolic dysfunction in order to successfully halt and prevent many chronic conditions. 

 

Learning objectives:

–    Understand that insulin resistance (IR) is a condition that precedes prediabetes
–    Learn how the challenges of measuring IR have been met and addressed
–    Discuss how metabolic conditions contribute to cardiovascular disease, chronic

       kidney disease, and nonalcoholic fatty liver disease
–    Learn the latest recommendations from the US Preventive Services Task Force on

       screening for prediabetes and type 2 diabetes

 

Time of talk: 60 minutes

Date:
Nov 09, 2022
Location:
virtual on-demand webinar
Presenter(s):
Michael McPhaul, MD, Senior Medical Director, Endocrinology, Quest Diagnostics

Hi, everyone, and thank you for joining today’s webinar on the early identification of Type 2 Diabetes risk in the primary care setting. My name is Trisha Winchester. I’m the manager of the Clinical and Education team at the Quest Cardiometabolic Center of Excellence at Cleveland Heart Lab. I’ll be introducing our speaker and moderating today’s question and answer session. First, I’ll cover a few items. You are in listen-only mode, so please ask your questions in the Q&A box and I’ll collect them and get to as many themes as we can. After the program, we’re recording today’s talk and we’ll be posting it on our website. So please, you can share it with colleagues or rewatch it for people who couldn’t join our program today. There is one AC last place credit available to those who complete the evaluation and a link after the webinar. And lastly, the views and opinions presented in today’s programs are those of our Speaker based on his professional experience and expertise. And so that leads us to introducing our speaker today, Dr. Michael McPhaul. He is the senior medical director for endocrinology at Quest Diagnostics. He prior to Quest, he was in practice for 25 years. His primary focus has been on developing tools to assist in the comprehensive care of patients with prediabetes, metabolic syndrome and diabetes. Dr. McPhaul, You can take it away. Thank you so much. Very nice to be here today and to talk to you about some of the things that I’ve been working on at Quest since I joined Quest about ten years ago after being on the faculty at Southwestern for 25 years. And this is sort of an overview of the talk I’m going to give today and talk about scope, the problem, current approaches to trying to prevent diabetes. Tools and applications to sort of improve that. And at the end, I’m going to talk about the application in a number of different areas, including risk prediction, the diagnosis of hepatic steatosis and following individuals through a diabetes prevention program. The first concept I want to point out is diabetes, disease, interception. And this is a concept that’s been sort of bandied about for about a decade or so. And the idea is, is that costs in terms of caring for patients, this is based on 2010 Medicare data is distributed non uniformly across the disease spectrum, where about 15% of beneficiaries actually account for 75% of the spending. Many, you know, very costly diseases have precursors and the idea is to intercept the disease before it gets to be the more costly versions. At a point in time where care is much less easy and much less expensive to deliver. I want to point out this is with reprinted with permission of Todd Pruitt and Adam Coffman from a seminar many years ago. The point is, is that the basic premise is to address the root causes of disease by intervening earlier than commonly accepted criteria for disease diagnosis and stop or potentially reverse the progression to more severe disease disease states before the individual becomes sick and becomes a costly burden for the health care system. So to intervene at these points now, at this point, I’m going to talk as a part of this, even though this is applicable to many other types of disease, whether we’re talking chronic kidney disease, whether we’re talking about cardiovascular disease, I’m going to talk about diabetes. And even though there are a variety of different flavors of diabetes, I’m really going to talk primarily in this discussion about type two diabetes, where there is a combination of both a resistance to the action of insulin, inadequate production of insulin and or both that actually causes the progression through insulin resistance to prediabetes and then on to diabetes. So I’m going to talk about this primarily because it represents the most frequent and largest burden of diabetes disease in our country. You know, this is something that you all have seen at various times. It’s just simply to point out that the level of disease and the degree of disease has changed with time. I’m not sure why this keeps advancing, but that’s not cool. The point is, is that in 20 in 1994, the most common prevalence was very low in terms of obesity and in the realm of diabetes burden. But by 2013, this had progressed dramatically to a much higher frequency of disease. Now, this is simply a depiction of the progression itself. And this is a broad and coarse generalization, but generally speaking, in any individual, as body weight increases, so does the degree of insulin resistance and eventually the requirement. Insulin resistance for increased levels of insulin resistance outstrips the capacity of the pancreatic islets secrete insulin to maintain normal homeostasis resulting first the appearance of prediabetes and diabetes. And this accounts for about 90% of the diabetes in the United States. And this is simply also depicted here the ways that hemoglobin A one see fasting glucose and fasting insulin changes during the spectrum. Initially when has normal fasting glucose and fasting and hemoglobin A1 C is normal with normal fasting insulins. As insulin resistance progresses, these two hemoglobin A1, C and fasting glucose remain normal, but fasting insulin actually rises to maintain this and prediabetes, this becomes a bit of a mismatch where the amount of insulin that is capable of being produced is inadequate to maintain either hemoglobin A1, C or fasting glucose or both in the normal range. And ultimately this outstrips the ability of the body to produce enough insulin and diabetes. And since So what is the scope of the problem? As you all may well know, 28.7 million people have been diagnosed with diabetes in the United States and eight and a half million don’t even know they have it. The care just put it in perspective. For those with known diabetes, the estimated about $237 billion. By contrast, the entire gross domestic product of Egypt and Finland are 235 and $252 billion a year. Prediabetes is about two and a half to three times more frequent. And of this number, 9/10 of them don’t know this and a substantial number of these people will go on to develop diabetes mellitus within five years. And then is this group that the pre diabetics that have been targeted in the US for intervention via the national program that’s been implemented by the CDC, the diabetes prevention programs as a result of it being something that is lower medical costs, not requiring expensive medications. And this is at a stage when it is really viewed as being potentially reversible. And I’m going to talk just a brief bit about diabetes prevention programs on the next slide. This is a slide that’s taken from the seminal paper that summarized the results of the diabetes prevention program in the United States. Intensive lifestyle intervention was designed to achieve and maintain a weight reduction of 7% of initial body weight through healthy diet, low fat diet, and to engage in physical, physical activity of moderate intensity, such as brisk walking for 150 minutes for a week. The magnitudes of reduction that were observed in this study were quite similar to those that have been observed in other studies in China and in Finland. And these observations establish the validity of such approaches and gave rise to this national diabetes prevention effort. And in this group has shown the placebo arm, metformin arm and the lifestyle modification arm, the lifestyle modification arm, resulting in about a 60% reduction and progression to diabetes in this group of pre diabetics at year four. That’s good. But what happened in the long term output of this? So this was summarized in a paper that was published in Lancet Diabetes and Endocrinology in 2015, and this is called the DPP poster or the Diabetes Prevention Program Outcome study. It looked at 88%, nearly 3000 of the surviving initial DPP cohort and it looked at it out to 15 years. The thing that I find remarkable about this is that even though diabetes incidence in this cohort long term was reduced by 27% in the Lifestyle Intervention group, an 18% in the metformin group compared to the placebo group, fully 55% of the people who were in the diabetes cohort that was treated by lifestyle intervention had progressed to type two diabetes. And I, I think that’s kind of sobering, really. I mean, this is our national program. This is one of the things we’ve implemented to try and prevent progression to diabetes. And in this arm at 15 years, 55% of the people have progressed to develop diabetes. So the diabetes prevention programs have been shown to be effective, but the effects are not permanent and the residual effects of the program appear to decrease over time. And these observations invite the consideration of potential modifications should we continue longer? Would such an extension be necessary for everyone? Should we start earlier? And if so, when? And more importantly, and I think this is the part that I’ll come back to at the end, is that at any point in time, how do you know where you are in the insulin resistance spectrum and these people with pre-diabetes, how do you know when you have ultimately succeeded and gotten them to a point where they are no longer at risk for the development of diabetes, cardiovascular disease or other diseases associated with insulin resistance in future? And this really requires a development of simple, facile and reliable markers of insulin resistance. And that’s what I’m going to talk about off and on for the remainder of this discussion. And I’m going to focus first on what I view as one of the really crowning achievements of Quest and the mass spectrometry department over the last decade. And that’s a high throughput, mass spectrometry based assay to measure intact insulin and c-peptide. Again, I’ve mentioned this spectrum before, and I want to focus in on this here, which is fasting insulin. And I’ll point out that this is informative at a stage and at stages before even getting to prediabetes. And one of the major problems here that we need in order to make use of this is reliable assays. Insulin and c-peptide assays have long been recognized as highly variable and in need of standardization. These are because these are for the most part, except for quest Immunoassays that are in fact going to be using different antibody pairs, different conditions of intubation and the like. And as a result of this, this variability between assays has hampered the ability to compare results across platforms and to employ these assays clinically over time. And particularly specifically, this variability has hampered the ability to use basal fasting measurements to assess insulin resistance. It’s impeded the definition of clinically relevant specific cut points for disease. And so even confused the ability to compare results between clinical and epidemiological studies that have used different methodologies and the recognition of these assays, the shortcomings of these assets has created the creation of international groups to effect the standardization of these assets and these and that highlights the number of peer reviewed publications, including the diabetes care. An example of this variability is taken from a clinical chemistry paper published in 2007, cited at the bottom in which they took patient pools and submitted them to a variety of laboratories running insulin assays as on a number of different platforms. And I would just invite you to peruse these and this is a low sample pool and a high sample pool, and you can see that there are fairly dramatic differences, particularly in the low sample pool from top to bottom there being as much as a 20 fold difference from highest to lowest, using the same sample pools and measured them via different insulin acids. So the objectives of what we set out to do is to get the ability to generate results that measure insulin by an intact tandem aspect and also to simultaneously quantify intact c-peptide. These are still as yet unrecognized. We have the potential to expand to human to insulin analogs and also to identify unusual variants. But we really haven’t expanded this capability to this point. I’m going to just really go very quickly to just simply point out that, you know, this is a two chain molecule. It starts out as a linear pro hormone that includes both insulin and c-peptide and the measurement and the commercial assay, the multiplexed assay in addition to a B chain specific mouse monoclonal immunoglobulin that’s included to enrich the preparation prior to measurement on mass spectrometry in those that are multiplex. Another antibody is also included. So this is an immuno enriched process. This is simply a flow of the overall progression to end up ultimately with the measurement of intact insulin and c-peptide on a tandem aspect, the quest intact insulin and C peptide assays are the first commercial available LCMS and assays for either analyte that measures only the intact biomolecules. These assets have been carefully standardized peptide mass measurements. If we had chosen to, we could have reported these as mass measurements, but we instead chose to maintain the familiarity of international units per ML that most clinicians are familiar with. But we do have very precise capacity to translate these into mass measurements. If we want to referee need to. These assays permit the definition of standardized enduring results that are directly convertible to mass units and will not change. And these assays will permit the definition of specific points and reference intervals. These assays will also permit the correlation of insulin and C peptide measurements with coral, the measurements such as the direct measurements of insulin resistance using glucose measure, glucose clamp measurements or steady state plasma glucose measurements, which is Dr. Reid then Jerry Ravens version of the clamp. So the first thing I’m going to talk about in detail is can insulin and c-peptide measurements be used to assess the levels of insulin resistance? And this was done in collaboration with Dr. Faheem Abbasi, Dave Shiffman, Carman, Tom, Jim Devlin, myself and Jerry Raven and Faheem and Jerry Raven were at the Stanford at Stanford University. These studies done. So this was based and this is published. I’ll show the citation in a minute on studies that have been conducted over quite a period of time with Stanford University. These are apparently healthy individuals without a history of cardiovascular disease. They were recruited from the San Francisco Bay Area through measurements, through advertisements in local newspapers. In the late nineties and early 2000s of participants with diabetes were excluded. And the study and the study that I’m going to talk to you about employed the fasting measurements that were obtained at the time. These individuals had the insulin suppression test performed at Stanford, at the Stanford General Clinical Research Center, insulin resistance for the purposes of this was defined as the top tertile of steady state plasma glucose in non-diabetic participants, and the analysis contained 535 individuals. These were selected from a larger group, but this group had a complete set of biochemical anthropometric measurements. This is the publication. It’s in the Journal of the Endocrine Society in 2018. And parenthetically, Dr. Revan is not listed as an author is acknowledged in the acknowledgment because he unfortunately passed away before he could view the final version of this manuscript. So we didn’t go into this with any kind of preconceived notions. And what we ended up doing was looking at the informative list of a variety of different potential covariance to inform the ability to diagnose and identify individuals with insulin resistance. And you can see that many of the usual suspects have very high p values when looked at and a univariate type of analysis in fasting glucose insulin c-peptide. The interesting thing was when you actually put these into a fully adjusted model, the only things that really remained significant were insulin and c-peptide. There’s a small amount of residual statistical value and creatinine and then BMI, but both insulin and c-peptide, when incorporated into the model, took pretty much everything else away from being statistically significant, which I found to be somewhat unexpected that both of these molecules were actually necessary because they are derived from a precursor proprotein. To put this in context and to put this in a little bit more detail, we did compare this to Homa. Are because this is something that people are very, very familiar with and looking at it either per standard deviation or top quartile versus not. Homa IRR does have the ability to discriminate individuals with insulin resistance and insulin alone and C-peptide alone do as well. But when combined into the model and this is the part that we’re going to talk about in terms of the insulin resistance risk score, this is a two and a half to three times better at predicting and identifying individuals accurately with insulin resistance as measured by the insulin suppression test slash SSP, PG as HOMA. So this is a better measurement. It does a better job of identifying individuals with important levels of insulin resistance compared to either insulin alone compared to c-peptide alone, or to whom are Now, I don’t really understand the biology of this at this point, but I will point out that the actual degradation and removal of both insulin and c-peptide differs. Liver is the primary point at which insulin degradation and inactivation occurs and excretion in the urine is one of the points at which C-peptide is primarily excreted. And you can see that generally speaking there are fairly good correlates high c-peptide which is shown on this axis, high insulin, which is shown on this axis, correlate with one another. But there are mismatches. There are some that are high c-peptide and intermediate insulin or vice versa. And I think that that is one of the reasons that this is more valuable is that this integrates signals from both insulin and c-peptide into a combined score and that this integrates information from both and sort of steps around some of the potential noise that occurs with either analyte relative to its metabolic rate. So the insulin resistance risk score has traceable levels of insulin c-peptide, which is used to calculate the risk. It reports the probability of having clinically significant insulin resistance to find as the top true type of insulin resistance measured by SS, BG and again using traceable and reproducible values of insulin and C peptide. And rather than reporting positive or negative, the actual reports themselves include measures that allow you to assess the relative level of insulin resistance. And the other implications is that insulin c-peptide are associated with insulin resistance. Independently, both insulin and c-peptide are required to adequately model insulin resistance formally assessed by SPG insulin and C-peptide can be used to assess the probability of insulin resistance in those with and without metabolic syndrome. I didn’t show this data, but what this was was derived from an assessment of those individuals in the Stanford cohort that had metabolic syndrome and those that did not. And then looking at which of those groups of individuals had insulin resistance using this methodology. And what we found is that the proportion of individuals overall in the group that was defined and applied, applying the score to those that had the full metabolic syndrome only accounted for approximately a third of the individuals with top tier insulin resistance, and the other two thirds was in the group without full blown metabolic syndrome. So it can be used to assess the probability of having high levels of insulin resistance, both in those with and without. And again, this risk score can be used to sort of provide a summary of the probability of having clinically significant insulin resistance. There are a number of potential applications of this, and I’m going to step through some of these throughout the remainder of the talk. But I am going to point out this in a kind of summary fashion on this slide. The identification of insulin resistance even before prediabetes is evident. And this is, as I said, shown in the validity of the presence and absence of the metabolic syndrome. And this is identified insulin resistance even before what most people would identify as clinically evident markers such as fasting glucose or hemoglobin A1 C aberrations are evident. It can also be used in repeated assessments of insulin resistance over time, for example, in the context of a response to intervention, for example, in the assessment of response to a DPP intervention, I’ll come back to that at the end. It can also be used to identify future risk of diabetes or future cardiovascular events. These are in population based studies. I’ll touch on these what we’ve done so far. And finally, it can also be used to identify individuals with hepatic steatosis. And this is using comparisons to those individuals with liver fat using magnetic resonance spectroscopy, which is considered to be the gold standard for noninvasive assessment of hepatic steatosis. The next three slide simply shows some of the basics about what this test looks like. This is the kind of depiction that you get if you order the cardio IQ Insulin resistance risk score gives you a readout of where you fall on the slider. You’re going to get a value from 1 to 100. The higher the number, the more likely you have significant insulin resistance. This is some details about the test code on the packaging, the transport temperature and the stability. This is a very well behaved test. It was developed in large part by Dr. Steven Taylor here at Quest. And I got to say, this is one of the most productive tests that I think has been recently developed. And we do a lot of good test development here. But this is one excellent test in terms of its mechanics and its stability and here are some related tests that may be of interest to individuals on this car. All right. So I’m going to step through in the next couple of slides, a discussion about insulin resistance and the risk of future disease. The first one I’m going to talk about first, I’m going to talk about some of the challenges we face because, you know, this is a new test. This is not the same as your immunoassays for insulin c-peptide. It’s based in mass spectrometry in order to do this, we actually have to have the availability of residual samples for such measurements and a period of sufficient time to actually discern and incident events. And I’m interested, most interestingly, most interested in talking with you about the future risk of type two diabetes, cardiovascular disease, stroke, not cause mortality. And we have a number of existing collaborations surrounding this, both with the Framingham study as well as Eric, as well as Milano. And I’m going to mention very briefly some of the results that we’ve published or that we’ve conducted in the both Framingham and the MALOMO study. So the first one I’m going to talk about is Framingham. Unfortunately for this one, residual samples were not available. So what we had to use and looking at this was the impact of insulin resistance in predicting future diabetes, using existing data from Framingham. I was actually kind of surprised that this hadn’t been modeled before, but I work very closely with one of the Framingham investigators, James Meggs, and we did this together and published this about two years ago in diabetes Care. We used the Framingham Heart study data and we excluded patients with type two diabetes, categorized them as either normal by hemoglobin, A1, C or pre-diabetic. We also grouped them into home with TERTILES. The follow up was for a median of around six years for incident diabetes, and these were adjusted for age and sex. The rates and counts of subjects were calculated for all 2200 study subjects and also for all 1583 normal glycemic subjects with fasting glucose greater than 100. The bottom line is here is that the elevated hemoglobin A1, C and Homa are independently predicted type two diabetes and all subject groups and even in normal glycemic subjects. This is a complicated figure that’s taken from the diabetes care manuscript that we published. This is the citation. At the bottom, what you see is the predicted probability of type two diabetes on the Y axis, and you see three groupings for insulin resistance. Tertile by Homa are Tertile one, which is the lowest, intermediate and highest, and then the values for the all of the subjects are depicted in black and those for fasting glucose. Less than 100 are shown in blue, and each of these tertiles is segmented into those with normal hemoglobin one C and those for prediabetic Hemoglobin one C. What you can see is that insulin resistance has a profound effect, whether it’s in the normal glycemic or in the non normal glycemic total population and predicting future diabetes. And this can be shown a little bit differently on the right hand side where you can see that the odds ratios for Tertile three versus Tertile one, whether it’s in the normal, the entire population are the normal glycemic population actually shows substantial clinical significance. The other one that I want to mention, we also published a paper looking at the predictive capacity of this test in the Memo Prevention Project, which is a large population based study that was taking place in southern Sweden. We published separately the predictive ability of this test to predict future diabetes. It also showed a very similar kind of result to what we saw in the Framingham study. But the thing that I wanted to point out here is independent of the actual risk of future diabetes, we also looked in this population at cardiovascular and all cause mortality in this, and we used this and uses the cutoff point and this greater than 80% and the probability risk score. And what we found was a fairly dramatic effect in showing about a 50% increase in predicting future cardiovascular disease and cardiovascular disease or all cause mortality in this population is I thought it was a fairly dramatic effect. This is as yet unpublished, but it will be published in a short and short order. The next thing I wanted to talk about is in the application of this tool for trying to define specific cut points and this particular work was done in collaboration with Fernando Borrell, who’s now at the University of Alabama Birmingham, as well as with Cancun. See his mentor until he moved to the University of Alabama, is at the University of Florida, Gainesville. And this was also done with a number of other collaborators at the University of Florida and at the University of Alabama for this study. It’s just simply to recognize that fatty liver disease is tightly linked to insulin resistance. No clinically available measure of insulin resistance has been able to predict fatty liver disease, in vitro measurement of plasma in serum, and previously had not been standardized and immunoassays detect variable amounts of other insulin fragments. And so what we did was conduct a study to determine whether measurements of the TAC molecules, insulin and intact c-peptide produced could predict the presence of fatty liver disease. And this was a patient population that was recruited from the general population, as well as the hepatology and endocrinology clinics at the University of Florida, Gainesville, and included individuals between the ages of 20 and 21 and 70. And these had patients and the study that we published that both individuals with and without type two diabetes, no other liver disease, and we excluded estriol to two inhibitors, GLP one agonists and thiazolidinediones agents as well. All patients underwent OJT tears, MRIs, liver biopsy if they had fatty liver and measurement of insulin and c-peptide. But by ten times black. I didn’t put in this. It’s in the paper, but I didn’t put it in this slide about type two diabetes, the insulin and c-peptide measurements were not particularly useful in identifying individuals with type two diabetes, simply because I think the prevalence of hepatic steatosis is high enough in patients with type two diabetes that it really doesn’t add much to the ability to discern those people with excess liver fat. By contrast, then fasting intact insulin significantly outperformed performed other clinical variables for the diagnosis of fatty liver disease. And this is particularly true when we combine that with LTE. We’re in this public session. We identified that it actually had a sensitivity and specificity that showed on this slide, which is actually quite good. This is the paper that we published that was in J.S. and was published about a year ago. And on this we show the results for the intact insulin and LP combination Nafld liver fat score, HCI and drug measure indexes as well. And the only point here is to simply say that this is a very effective way to discern in a nondiabetic population those individuals who have a high probability of having fatty liver disease when compared to the gold standard of magnetic resonance imaging. So fasting insulin measured by tandem aspects predicts fatty liver disease when measured by r a in patients with diabetes. This was contained in the paper. No insulin measurement was really useful to predict fatty liver disease and type two diabetics. And in fact our algorithm for the evaluate of individuals with fatty liver disease immediately emerged to trying to identify individuals with significant fibrosis and those that are known to have type two diabetes, fasting liver insulin predicted fatty liver better than levels that were measured during OJT, and the combination of fasting, insulin and plasma alt for the diagnosis of fatty liver disease outperformed a number of other well validated studies and scores. This reinforces, in my view, the utility of using standardized insulin measurements to assist in the diagnosis and management of insulin resistance related conditions. And the other thing that I just mentioned is sort of parenthetically, we haven’t done this yet, but I’m intending to have built into our existing insulin resistance risk or a flag notification that is triggered by having insulin level that is measured as greater than ten International micro international units per IML, as a way of notifying clinicians that their non-diabetic patient, if it is a non-diabetic patient, has in fact a high probability of having a paddock steatosis and potentially may need further evaluation or risk stratification for fatty liver disease. Okay. So the next little bit that I’m going to talk about here is really a almost just a vignette. And again, it sort of builds on the theme of notifying the clinician. We measure a lot of tests looking at lipoprotein fractionation to identify patients who have residual risk beyond simply the measurement of total or LDL cholesterol. And this is used quite widely across the clinical practices of clinicians in the United States. This was done in combination with a number of collaborators. It’s built on an existing collaboration, as I mentioned, with Faheem Abbasi and Josh Knowles, who are both very close colleagues with Jerry Raven at Stanford. It also included contributions from a number of people, including Charlie Roland and Dave Shiffman, who are consultants currently. There were prior employees at Quest Diagnostics, as well as Michael Caulfield, who was a senior scientific director at Quest Diagnostics for many years. And all we did here was take what we knew about the samples that we had already measured species on. And this was actually in part the group that led to the definition of the insulin resistance. And we measured iron mobility fractions across the entire spectrum of disease and specific type of practice lipoprotein fractions have been shown to have characteristic changes in insulin resistant individuals, particularly, for example, the abundance of small dense LDL. And Dr. Charlie Roland developed a way called the LCR score, which integrates this information. This basically takes all of the different fractions that are associated significantly with insulin resistance, integrates them into this overall score. And all of this shows and I’m not going to go into the details. This is unpublished that’s under review at the Journal of Preventative Cardiology. Is that alone or in combination with other variables? This has a very, very good way of calling out individuals that have potentially high levels of insulin resistance. And again, as I mentioned, with the insulin resistance score and the insulin above 10.5 micro international units per animal and predicting hepatic steatosis, it’s my intention to try to add to the ion mobility offering a trigger off of the score that actually potentially will cause individuals to be alerted to the presence of potential significant insulin resistance in individuals who they may not otherwise suspect are insulin resistance. Okay, So in the last segment of this, I’m going to turn to a the use of measures of insulin resistance in a longitudinal context. And this is going to describe a study that we’ve conducted, haven’t published yet. It has the incredibly imaginative name of quest testing to assess insulin resistance for Q tear. And I take full, responsible responsibility for this incredibly creative name. So a number of studies to recap, including the Diabetes Prevention program, have demonstrated the utility of programs to delay or prevent progression of the individual’s prediabetes to type two diabetes. And as I mentioned, despite these successes, a substantial number in the long term depressed, 55% and lifestyle group developed type two diabetes. And the reasons for this persistent progression really remains undefined. And we at Quest hypothesized that the diabetes prevention programs might conceivably represent interventions that represent too late and too little. For example, they may be efforts initiated a period when metabolic derangements have already evolved to produce significant discernible abnormalities in glycemia and circumstances where a pre formatted program without an integral mechanism to formally identify residual risk at the conclusion of the program to identify those individuals that might benefit from continued intervention. And I also point out that this latter aspect might also conceivably create an appropriate sense that metabolic risk had been negated in individuals, where metabolic risk still remains. So to approach these issues, we developed this, as I said, this mass spectrometry assay and demonstrated that the past combination of fasting insulin c-peptide measurements performed on a single fasting measurements permitted the inference of whole body insulin resistance as assessed by the insulin suppression test and form research method employed to quantitate insulin resistance. So in the current study I’m about to describe, we use this insulin resistance to risk score to examine the attempt, the initial patterns of insulin resistance in individuals participating in for DPP and to follow the evolution occurring in these measurements in participants during the program. This permitted us to explore how repeated measures to examine the changes in insulin resistance in the course of the DPP or any other kind of intervention for that matter, could classify participants either initially, during and at the conclusion of the Diabetes Prevention Program. And this is my own personal view here. I feel the findings suggest that such assessments can subdivide individuals with similar clinical appearances into distinct groups, and that the assessment of insulin resistance in participants can provide both a useful method for interim feedback and identify individuals and whom additional intervention may be beneficial. So on this is described is a convenience cohort. This convenient corre was not acknowledge them as we go through. This was affected by two of my internal collaborators here at Quest Diagnostics, Martha Brandon and Julie Larson. These were CDC certified DP’s and they were really being explored at the time because we weren’t sure whether Quest was going to offer diabetes prevention programs as a part of its offerings to clinicians. So I took rank advantage of this. And basically, in addition to all of this, we got RB consent to obtain blood samples. Anthropometric at baseline, one month, three months, six months and 12 months. What I’m going to show you is the results of some of these findings. So what you see here is to the left, subject number, initial classification, diabetes, pre-diabetes are normal. Some of these people got into this because they were included and enrolled on the basis of historic values. I also have here this kind of cryptic initial and final glycemic category. Patients in this had either glucose or hemoglobin A1 C that classified them as either diabetic, pre-diabetic or normal. And these were the initial group fasting glucose, initial hemoglobin, A1 C measurements and what they classified them as, and then their final glucose or hemoglobin A1, C classifications. You see their initial insulin resistance category and then their final initial excuse me, their final insulin resistance category, their beginning and final weight, their percent weight loss, their beginning fasting glucose and beginning and fasting front of glucose, their beginning and fasting final excuse me beginning and final hemoglobin A1 C and their beginning and final insulin resistance risk score. Now I focus this particular slide just on those that had an initial insulin resistant category of high. Now we’re about these two fellows down here first because they really didn’t take this seriously. This individual lost zero weight. They were the same at the beginning, at the end, although they came back from their lab tests, this individual did the same thing, but actually gained two and a half percent. They went from 147 to 153. So it’s not to surprising that their insulin resistance risk didn’t change. These two individuals had high insulin resistance risk scores at the beginning. You can say 99 and 70. They both lost substantial amounts of weight. I can see that here. They this is reflected in the normalization of one to a normal fasting glucose and near normalization of their hemoglobin AA 1c1 with an improvement in their fasting glucose and then in the near normalization of their hemoglobin in one C. And both of those had very good drops in their insulin resistance risk for these individuals. These two both lost substantial amounts of weight. They didn’t do quite as good in terms of normalizing their fasting glucose or their hemoglobin one C, but both had drops in their insulin resistance risk. For these individuals I find to be just nothing short of amazing. These individuals had all of them significant amounts of weight loss, some of them very dramatic amounts of weight looked at weight loss. But all of these beginning to end had still elevations of their insulin resistance risk or and I want to point out for people in particular, and that these individuals normalized their fasting glucose, they normalized their hemoglobin a1c all for the red arrows. Yet they all have insulin resistance risk scores that are pegged at the top end of the spectrum. I would point out that in that Malomo reference that I pointed out, these individuals are above the threshold that would be associated with a 50% increase in incident, cardiovascular risk and all cause mortality in future. All right. So, you know, again, I’m trying to not get too dramatic about this, but I think that the pattern observed in the participants with high initial insulin resistance risk course was provocative. Although weight loss was observed in all patients, high insulin resistance was observed in a high proportion about half of those individuals At the end of the observation period. This pattern was even encountered in those individuals that met and exceeded the DPI goals of 7% Weight loss. This last observation is particularly interested in light of the linkage of elevated insulin resistance to the future development of diabetes and even in individuals with normal glycemia. This suggests to me that the high levels of insulin resistance may persist in a significant number of subjects at the conclusion of the successful participation in a deep even an individual’s obtaining, attaining substantial weight loss and normal fasting glucose and hemoglobin a1 c at the end. And to me, this raises the possibility that a formal assessment of diabetes prevention program participants at the conclusion of a diabetes prevention program might assist in the identification of those individuals at risk and in need of additional intervention and or weight loss. So just to sort of summarize, classes create an assay to measure intact insulin and c-peptide, but then a mass spectrometry and we’ve talked about it’s used to model the measures of whole body insulin resistance as assessed on using insulin suppression test by Jerry Revan and colleagues at Stanford. And I’ve actually talked a little bit about the use of measures of insulin resistance to predict future diabetes cardiovascular disease and all cause mortality to identify hepatic ptosis by magnetic resonance spectroscopy or to identify individuals using noninvasive blood based measures in a way that approximates the ability to identify them using a paddock’s data by using magnetic resonance spectroscopy in nondiabetic subjects and to follow individuals through the course of a diabetes prevention program. And on the last slide, I want to acknowledge the fact that I’ve had the pleasure and privilege to work with a number of individuals in doing these things internally. I did not include, although I mentioned him, Steve Taylor, who actually was the creator or the principal creator of the intact insulin and c-peptide assays by mass spectrometry. He’s still with Quest and is a very important member in the Publications committee now externally. James Meggs at the MGA for him, Abbasi at Stanford, Josh Knowles at Stanford, the late Gerry Read from Stanford. Ken Cuzzi from the University of Florida, Gainesville. Fernando Brill, who’s currently at the University of Alabama, and Olema Lander, who is in Melman, Sweden. Dov Shiffman, who is now a consultant with Quest. Judy Lui, who’s a very talented statistician at Quest. Charlie Roland and Mike Coffield are both prior employees of Quest that are now consultants with Quest. Martha Brandon and Julia Latham are both very important colleagues of mine who I still work with closely and who are still with Quest. And with that I will stop and be happy, answer any questions that I’m able to answer. Great. Thank you so much, Dr. McFall. That was really wonderful information and really highlighting a lot of the work that Quest has done in terms of bringing this important assay to commercial to be commercially available for our patients, to really identify this unique window of opportunity, to identify insulin resistance before you see that changes in passing glucose in anyone. See, so the first question just starting with the basics was a question about clarity on the home air. Now we know this is a measurement of fasting, insulin and glucose that indicates insulin resistance. Can you kind of dive into a little bit about why why we suspect the insulin resistance panel score outperforms that? Well, there’s no, you know, guessing. I mean, it’s pretty well clearly established in the presentation that I showed. And that’s summarized in the paper that was published in the Journal of the Endocrine Society. I can’t explain why. I mean, if you look at the odds ratio for insulin versus, oh, they’re very similar. And if you look at the odds ratio for C-peptide compared to home air, it’s a little bit better, but it’s approximately the same. The thing that comes into play here is that the combination of both and I got to say, when I did this analysis, I didn’t start out trying to prove insulin and C-peptide was in fact better than insulin or c-peptide alone or better than homa-ir. That’s what the answer was. And as I said in discussing this, I don’t know, but I suspect that it may have to do with the fact that C-peptide and insulin both have distinct metabolic fates that may be evident more or less in one individual versus another, and that by looking at it from the perspective of integrating information from insulin and c-peptide into a single algorithm, that this captures maximal information relating to the markers, insulin and c-peptide, which are both independently associated with the measured levels of insulin resistance in the insulin suppression. So I think it’s a great question. I don’t know the answer, but that is the observed results and the odds ratios as I said are about two and a half to three times for insulin and c-peptide combined combined. Compare it to i r whether you look at it from top quartile versus not or whether you look at it per standard deviation. So it’s it’s not and this is a big data set of 35 people. So it’s really not something that is marginal in terms of its statistical significance. Yeah, thank you for that. And kind of to expand upon how it compares to other measures of insulin resistance, a question was asked in the chat or in the Q&A excuse me how the panel will score. Have we done any insights into see how it compares to a one or two hour oral glucose tolerance test? No, that’s a really, really great question because I just want to make sure it’s clear that the modeling that was done in the Stanford study was fasting insulin and c-peptide compared to measures that assess whole body insulin resistance. So even though they approximate one another, you know, they depend on, you know, that that relationship is is very specific. All of the studies that we’ve done looking at future risk of type two diabetes, hepatic steatosis of future cardiovascular disease, all of those are still based on fasting measurements. But it’s quite clear that there is a subset of individuals that are displaying primarily or in combination with the whole body insulin resistance, impairments of insulin secretion, particularly as it pertains to a glucose load or a dietary low. So I think that if one looks at the insulin resistance risk score and it lines up with the glycemic measures that you see those individuals can be followed, I think with the insulin resistance risk score. But there are individuals who have, you know, aberrations of glycemia that are not displaying high levels of insulin resistance. There are individuals like that. And my guess is, is that those individuals likely have such a significant impairment of insulin secretion that it may actually cause their insulin resistance risk score to be somewhat lower and those individuals probably should have another GTP in order to identify residual risk. Yeah, it’s an IT that’s a really interesting concept that you really dived into. I wanted to like in the most general sense though, so kind of taking a step back, there’s been a I thank you all for the questions in the chat. This is very great. But just taking a step back in terms of an agency is an average of fasting glucose over a three month period. The insulin resistance panel score, as you mentioned in, you know, multiple different avenues that this is really identifying that risk earlier on. And so that’s kind of why that question had arisen. There was a couple other questions that had come up around that. So thank you for that, that detail explanation. Another question about the populations that have been studied, how has the panel score been assessed in pregnant women and what is it in pregnancy that triggers gestational diabetes? Oh That’s a great question. Actually. This is like a hit list Reading from my my future plans know that I think the things that I will tell you that I right now I have zero data on a normal pregnancy or individual developing gestational diabetes. And I’m very interested in establishing cohorts to study that. I would take a very large bet that if you had pre-pregnancy insulin resistance risk scores that you would identify and enriched a patient population, that would be enriched for those that would go on to develop gestational diabetes. I can’t prove that yet, but that’s my baseline assumption. It may not get everybody, but it’s going to identify a subset. I think that would go on to develop gestational diabetes. And as I said, I’m really interested and my email address is all over the Internet, so I’m pretty easy to find. So if anybody has additional ideas about patient cohorts, I would think that a patient population that’s based in individuals who are struggling with fertility, such as patients with PCOS or in a fertility clinic who have tubal disease or other impediments to the normal fertility that they might otherwise have, would be good individuals to, you know, get a sense of how insulin resistant they are and then follow them through pregnancy. We have established a number of RB consented protocols here for the purposes of doing, you know, for for example, the cuter study that I just mentioned. And I’m thinking that that would be something that would be very interesting to conduct with external collaborators as well. So I think that that is an open question. I think it’s a very interesting open question, and I can see any number of different things that might feed into putting at risk for developing gestational diabetes. But that’s a very long way of answering and saying, I don’t know yet. I mean, you’ve done a lot of great work thus far, so I can only imagine what you’re going to do in the future. And I think there’s a lot of excitement about what this test can really unravel. So thank you for all your hard work on that. I know we’re at the top of the hour, so I want to just thank you for your time and in providing this talk. I think there’s a lot of really great information. Thank you to all of our attendees. There will be a link sent out in two weeks for the recording. If you would like to re review it. I believe that there is a link in the chat for a web page with more information and so feel free to to join that. And thanks again, Dr. McFall. Any closing comments that you’d like to end with now? Thank you very much for your attendance. All right. Have a great day, everyone. Thank you. Bye bye.

Page Published: January 09, 2023