Fourth Annual Friedreich's Ataxia Symposium

Transcript: Diabetes in Friedreich's Ataxia

Transcripts in This Video Series
Neurological Features of Friedreich's Ataxia Diabetes in Friedreich's Ataxia Biochemical Mechanisms and Pathways Questions and Answers Session - Clinical Panel Research Topics and Discussion

Printer Friendly Version

Steven M. Willi, MD: Thanks. And thank you for the invitation to come and talk. I'm not going to put myself up here as an expert in Friedreich's ataxia, David is, and that's, and I happened to be the diabetes person at Children's Hospital when we were conducting this trial that the FDA had requested that we look at the diabetes of Friedreich's ataxia. And so I've been on the learning curve for the past couple of years.

So I want to share with you this morning some definitions. We're going to start talking a little bit about diabetes and we're going to have to cover a few definitions so that we can understand diabetes and also talk about some of the pathogenesis or the origins of common forms of diabetes. This is so that we can get a background and understand how Friedreich's ataxia can cause diabetes and the origins of diabetes in Friedreich's ataxia. Then I'd like to touch upon how we can assess the risk of diabetes. That's so that I can present the information that we have from our clinical trial and so that it can be understood, and hopefully you'll get an appreciation for what we learned from that trial. As David mentioned, the results were not significant, unfortunately, but I can tell you a little bit about what we learned from the trial anyway. And I'll present some of those results and a little bit of a summary perhaps.

Okay. So let's start with the basic definition of diabetes. Diabetes mellitus, what we call diabetes mellitus, is a lifelong condition. It's marked by high blood sugar, and it's caused by either too little insulin in the body, or an inability of that insulin to work, or a resistance to the insulin, or both, or a combination of both. Now, we have diagnostic criteria that we use to define diabetes, and the symptoms of diabetes with a blood sugar of greater than 200 is sufficient to cause, to lead to a diagnosis.

The symptoms, classic symptoms of diabetes are polyuria or frequent urination, polydipsia or frequent thirst, an excessive thirst, unexplained weight loss, and there are a number of other potential symptoms, types of infections and such, that can be sort of typically associated with diabetes. We also have some laboratory criteria, strict laboratory criteria, that can lead to a diagnosis in the absence of any true symptoms, and this has developed over the years and probably has helped to define what we describe in Friedreich's ataxia and diabetes. What I mean by that is that a single fasting blood sugar of greater than or equal to 126 is sufficient for a diagnosis of diabetes. Typically, people with a blood sugar that's under 200 or so won't have any symptoms of polyuria and polydipsia and such.

We also have something called a hemoglobin A1C that can be measured and actually can be measured from a single drop of blood on a finger stick, and is actually being done now for screening for diabetes quite frequently. And then there is also a diagnostic criteria that we rely less and less upon which is a two-hour blood sugar greater than 200 after something called an oral glucose tolerance test. And you may not be able to see the cartoon over here, but there's a doctor and there's the Energizer Bunny sitting on the exam table and he's saying, the doctor's saying, "So you say you have to keep going, and going, and going. We better check you for diabetes." So, it's alluding to the frequent urination, I guess.

So, if we look at a classification of diabetes and how diabetes presents to us, it comes into three major types, and then there's sort of a hodgepodge of other minor types that I won't go into any detail about, but it's important to look at, particularly the first two types, Type I and Type II diabetes, because they are very different conditions, and there are elements of each of them in the diabetes of Friedreich's ataxia, so we'll start with an understanding of them and move forward.

Type I diabetes used to be called insulin-dependent diabetes or juvenile-onset diabetes. Type II diabetes is often referred to as non-insulin dependent or maturity-onset diabetes. Gestational diabetes is a special form of diabetes that basically is defined as any diabetes that has its onset during pregnancy. It could be Type I, it could be Type II, if it starts during pregnancy, it's called gestational diabetes. And then, this hodgepodge of other specific types. There are genetic forms, one called maturity-onset diabetes in the young, which looks very different from the Type I diabetes that typically affects young people. And, there are other forms of diabetes that are secondary to other conditions. There can be other endocrine conditions, there are drug induced forms of diabetes. Certain infections can lead to diabetes. And then there are a number of genetic syndromes that are associated with forms of diabetes. And then, you can have a condition like cystic fibrosis that affects the pancreas and just leads to a lack of insulin production because the pancreas is damaged.

So if you look at the pathogenesis or the origins of Type I diabetes, this is a chronic autoimmune condition, autoimmune meaning that the immune system attacks a natural part of the body. It clearly does result from a genetic predisposition, but there is some environmental trigger that we don't understand very well. And that causes an infiltration of this immune system cells into the pancreas. It's marked by the presence of something in the blood called an autoantibody. This is a substance in the blood that basically also aids those cells in attacking the pancreas. And there's ultimately destruction of the beta cells, or the cells in the pancreas that are specifically making insulin. There are several different types of cells in the pancreas, and less than 10% of them are actually those beta cells, but the beta cells are the only ones that are affected in Type I diabetes.

There are other autoimmune conditions that also can accompany Type I diabetes. There can be an attack on the thyroid, the adrenal gland, and the GI tract, which are also associated with Type I diabetes. If you look at Type I diabetes over time, what you can see, there's time, it's actually on the bottom, and the beta cell mass or this mass of cells in the pancreas producing insulin is seen on the y-axis. And the genetic predisposition dates from birth. I'm losing my pointer here, but hopefully you can see the genetic at-risk potential starts at birth, and then what happens is that there's an inflammation in the beta cells in the pancreas that cause multiple antibodies to be positive and a decline in those cells in the pancreas. There can be a loss of what we call first-phase insulin response, or the early insulin response that occurs, and then it eventually results in diabetes.

But, as you can see there, you really have to have only about 20% of your pancreas cells left for the diabetes to develop, and that's important because we have this abundance of cells that are there that are functioning that are keeping up. And, only at a point when we get beyond that 20% or so that we need do the symptoms of diabetes develop.

Now, the origins of Type II diabetes are somewhat, are quite different. There's several organs that affect Type II diabetes, that cause Type II diabetes to develop, and it's important to understand this because they are relevant to Friedreich's ataxia. In particular, if you can see, I can't point to it unfortunately, but the insulin, the pancreas is also involved in Type II diabetes, and is fairly important in Type II diabetes, but I'd like to focus on this part of the slide which is supposed to be a muscle cell or a muscle itself, and the muscle is really important in Friedreich's ataxia, as Dave sort of alluded to a little bit. And, as near as we can tell in the diabetes of Friedreich's ataxia, the muscles are also important because there's an impairment in the ability of the muscles to take glucose out of the bloodstream and into the cells to be metabolized for energy. And because there's this inability of the cells to take up glucose, it remains in the bloodstream and diabetes is the result.

So, in Type II diabetes, the liver also plays a key role, but that's not believed to be the case in Friedreich's ataxia, although that's not been specifically examined.

So we're going to focus, really, our discussion on what's going on in the pancreas, and what's going on in the muscle cell, and hopefully we'll get to understand how those key players are contributing to the diabetes in Friedreich's ataxia.

So to remind you, Type I diabetes is an absolute insulin deficiency. It typically has an abrupt onset of high blood sugars, and a propensity for developing a severe condition called DKA, or diabetic ketoacidosis. It requires insulin for life, and it's the predominant form that occurs in people under the age of about 30. Whereas, Type II diabetes is predominantly insulin resistance with a relative inability of the pancreas to keep up. There's a progressive decline in this beta cell function, but it occurs slowly over time and the hyperglycemia that develops, develops gradually, so it's initially very silent, and will come out in testing before it will come out in symptoms. It can be treated with oral medications, a number of different oral medications, can be used to treat Type II diabetes, which is not an option for Type I. And, an important part of Type II diabetes is that it's associated with the passage of time and the increasing of age, and is also associated with obesity.

In Friedreich's ataxia, the diabetes of Friedreich's ataxia lies somewhere in between here. And, going to present a little bit of information that may shed some light on that. Friedreich's ataxia, as David mentioned, is predominantly a neurologic condition. It's in fact the most common form of inherited ataxia, and it primarily affects the central nervous system, the spinal cord, the peripheral nerves as Dave mentioned. It also affects the heart and the pancreas, and the major manifestations, I'm just going to touch upon the diabetes aspect, but I think there are other speakers that are going to go into some of the other conditions. Genetics of it are known to be autosomal recessive in most cases, and in, it's in most cases it's due to a loss of function mutation in a gene called the frataxin gene. This gene causes the result, the mutations result from an unstable expansion of the gene, a certain portion of the gene has excess GAA repeats, and that's going to be important to understand from some of the data that I'm going to present to you. The more of these GAA repeats you have, generally the more profound the affect of FA. The number of repeats can vary quite a bit, from 66 to 1700 in reports, and compared to 7 to 34 so that it is occurring in people without FA.

And the pathogenesis as it relates to diabetes appears to be that this genetic condition leads to the production of abnormal frataxin, which then is incorporated into the mitochondria and causes an inability of the mitochondria to get rid of iron. And this accumulation of iron leads to a failure, really, of the mitochondria to function in certain tissues. And the details of that are really not that important, but what is important to diabetes is that mitochondria are found in very, very high concentrations in muscle cells, particularly the heart muscle, but all of the muscles of the body. And remember, muscles are vital tissue to diabetes. They are also found in very, very high quantities within just those beta cells of the pancreas, so it seems as though those beta cells are special in that they require a lot of energy, and the inability of the mitochondria to produce that energy may be the source, one of the sources, of the diabetes that develops in Friedreich's's ataxia. Of course, the central nervous system actually has more mitochondria per cell volume than almost any other tissue, so that's why the nervous system's affected.

So what about diabetes in Friedreich's ataxia? I've talked mostly about diabetes, a little bit about Friedreich's ataxia, and now what about the two together? So it was recognized many years ago that there's a high incidence of diabetes in Friedreich's ataxia. And, one study from the mid-1990s showed actually that either overt diabetes or impaired glucose tolerance, which is sort of a form of, a mild form of diabetes, occurs in approximately a third of the people described in that paper, there were over a hundred people with FA reported in that journal article. And when they tested people, they found actually that about a third of them had either diabetes or a mild form of diabetes.

What about the pathogenesis? How does the diabetes result in FA? Well, both the insulin resistance and the insulin secretion by the beta cells can contribute to the development of diabetes I've mentioned to you, and it is clear that impairment of either of these pathways can be attributed to mitochondrial dysfunction, and I've mentioned that the mitochondria, the part of the cell that produce energy, are the part of the cell that's affected by Friedreich's ataxia, so it sort of makes sense, and previous studies have actually demonstrated or suggested that perhaps both insulin resistance and beta cell function, or the interplay that occurs between those two properties are responsible for the blood sugar abnormalities in Friedreich's ataxia. And that was reported back in actually, a paper back in 1988. Not exquisitely well-done studies, but nonetheless did suggest that both the insulin deficiency and the lack of insulin action is playing a part in FA.

So when we in the field of diabetes are trying to evaluate, well, what's going on in diabetes? Is it a problem with the cells not taking up glucose in response to insulin, or is it a problem with the insulin producing cells not producing insulin? We have several tests that we do and I'm just going to go through them very briefly because understanding them is important to understanding some of the results of the trial that I'm going to mention. Actually, I'm not going to present trial results per se, but I'm going to present what we learned from doing the tests in individuals. And, the fasting, we can do a single fasting glucose and insulin, that's done without a meal on board, and what the body does is sort of finds an equilibrium. And, from that equilibrium that's established, we have mathematical models that have been developed that sort of can tell us from how high or low the blood sugar is from how high or low the insulin value is, and, more importantly, how they're related to one another. We can calculate what's the contribution of the beta cells or the muscle cells to the diabetes. We call those models the, there's the HOMA model, there's something called the QUICKIE model, and they're crude measurements, but measurements nonetheless. They can kind of answer that question.

We can also do the same thing with the oral glucose tolerance test. Many of you may be familiar with the oral glucose tolerance test. It's basically taking a blood sample and a series of blood samples as you're, before and after you drink a very sweet drink, and it presents a challenge to the body, and we can calculate something called the whole-body insulin sensitivity index from that. And there's something called an insulinogenic index that we can calculate from that. Those measurements are not as precise, they are somewhat variable because the delivery of the glucose is somewhat variable.

If we really want to answer this question exquisitely well, we do something called the intravenous glucose tolerance test, or an IVGTT. And any of the audience that participated in our trial will know what an IV glucose tolerance test is. It's placing an IV and giving a bolus or a load of glucose, allowing the blood sugar to climb, and then giving a dose of insulin to see how responsive the cells are to, the body is to that insulin. So we use that to calculate several things, one called the insulin sensitivity index. We use it also to calculate the acute response, insulin response to glucose. So we look and see what the blood sugar does in response to these various challenges, and then use mathematical equations to derive numbers that have been very well validated that sort of reflect and can be compared among populations to determine what the relative contribution of insulin resistance versus insulin deficiency is.

And then, there's something called the disposition index that I'm going to talk a little bit more about later that is very helpful in trying to determine where along the path toward diabetes a person might be. And it seems as though everybody sort of has a different path that is largely genetically determined, but may be determined by certain environmental factors as well. And that's what we tried to use in the trial of A000 to try to determine whether the medication had an effect on glucose. And it was the FDA that actually mandated that we have a glucose measure, because I think they were concerned about, you know, the neurologic measures not being as precise or consistent because of all the things that Dave mentioned earlier.

So this is what an oral glucose tolerance test sort of looks like, if you look at blood sugar, plasma glucose, and plasma insulin concentrations, and then at time zero you drink a drink that has a lot of sugar in it, you can see that in the people, in the people without diabetes, the blood sugar rises sort of gently and then kind of returns to a baseline within about two hours. In people with diabetes, it rises more drastically and then just sort of stays up there. In what you see in people without diabetes is that the insulin response is quite brisk, and then tails off as the blood sugar comes back to normal. In people with diabetes, typically this is Type II diabetes, typically what you'll see is an inappropriately low response compared to the high glucose level. So that's an oral glucose tolerance test.

This is a slide. It's very busy and I apologize for it, but I really just want to make a few points about what the IV glucose tolerance test is. And why it is that we do it, and what kind of information we get from it. This is again on a graph, a graph of time versus the measurements of the insulin or the glucose. We have glucose and insulin sort of on the same axis here, and you may not be able to appreciate it, but now that I've got a pointer, I can point out to you that there is a, if you look at the green tracing, which is what the insulin level is, and you look at the blue tracing, which is the glucose level, you can see that at time zero when we give a bolus of glucose. Okay. So we're pushing it in an IV, so it's not being, you know, drank and absorbed slowly over time, it's really almost instantaneous within a minute or two minutes that glucose is in there and you're measuring it. What happens is this blue curve goes, shoots straight up, so high and so fast that you really can't even see the line, but it's up here, and then it gradually comes down because the body is producing insulin and causing the blood sugar to come down. That insulin response is here in green, and you can see that the body pushes out some insulin immediately. That's called the first phase of insulin release, and if you've got healthy beta cells you get a nice big blip like that. If you don't have healthy beta cells, you get a flat line, or something in between. Okay? So we use that, we look at that first curve as sort of how healthy are the beta cells? How well are they responding? Okay? And then we look, then we give insulin. Okay? So, in the case where people don't make insulin well, we have to give them insulin because they won't, the blood sugar won't fall otherwise, and we have to have the blood sugar fall. So we give a dose of insulin that again causes a rapid rise in the insulin level and then a gradual decline as the body absorbs the insulin. And what's important is we look at the insulin levels and look at the rate of this fall in glucose. So if somebody's sensitive to insulin, that rate of fall will be rapid. If somebody's insensitive, it will be slower. And we use a mathematical model, so this is sort of the, it's in here a yellow box, you know, but I call it a black box. You take all these numbers, you run them into a computer that does a lot of calculations, and it spits out some information about what the beta cell's response is, and what the insulin sensitivity is. Okay?

So that's, it's detailed, and I apologize for the detail, but it's important to understand why we do this measure, and what it is because it gives us insight into the diabetes of Friedreich's ataxia, because it's still not agreed upon to what degree FA is an inability of the pancreas to produce insulin versus a problem with the cells not absorbing insulin and utilizing insulin to absorb glucose. Okay?

So a couple of more definitions here that are important to this IV glucose tolerance test is this acute insulin response to glucose. Okay? And that's again to remind you, that's that green blip in the front. Okay? That's the acute insulin response. Then there's this sensitivity index. Okay? That quantifies the capacity of insulin to promote glucose uptake. Okay? So that's the number that's essentially calculated from this rate of fall in the blue line with insulin around. Okay? And then, there's something called the disposition index, which is actually the product, or the multiplication of the acute insulin response to glucose, times the sensitivity index. Okay? And we add, we multiply these numbers to give us an index that tells us what the relative contributions of insulin, deficiency in insulin insensitivity are. Okay? So both of these factors, you can have one that's poor, or you can have the other that's poor, your number's going to be low. If you have both that are poor, your number's going to be really low. Okay? And a low number means a risk of developing diabetes in the future. Okay? So it's sort of used as a predictor for the future development of diabetes. Okay?

But it is a complex concept, so it's not really straightforward. I can't tell you, "Okay, I've got a number of 12 and that's good, bad, or indifferent." It's more complex than that. It's complex if you try to just multiply them out. We try to simplify it by putting it on a graph, so what we do is we put the acute insulin response on this side, and we put the insulin sensitivity index on this side, and then we actually have normal curves that people sort of follow throughout life. So as you get older, unfortunately, one of the things that happens is that your insulin sensitivity declines, so you're going to go down this curve, as you gain weight, you become more obese, you're going to go down this way. Okay, in general, from the right side to the left side, your insulin sensitivity is going to decrease. Okay?

A good pattern is that as your insulin sensitivity decreases, your beta cells, your pancreas responds appropriately, and you slide up this upper curve. Okay? But, if you're predisposed to diabetes, you have a curve that's a little flatter. Your inability to produce enough insulin is uncovered. Okay? And that's what disposition index basically tells us. It puts a point on this curve and tells you where you lie. Okay? And we did that with 31 people that went through the A000 trial. So we, and we've actually got a slide that'll show you where they fall compared to a normal population. Okay?

So disposition index kind of represents the ability of the beta cells to compensate for insulin resistance. It has a heritability of about 50% in many populations, so it's something that you pass along in your genetic makeup. It can predict the future development of Type II diabetes and it's linked to certain genes that are linked to Type II diabetes, so it sort of all makes sense and it sort of expresses your innate tendency to develop diabetes, and therefore might be a helpful measure for people with Friedreich's ataxia.

So again, if we look at people who are non-progressors to diabetes, what happens is that with the passage of time they don't fall further down and to the left, they actually just slide right up one of these curves, whereas, people who have a predisposition to diabetes over time will actually start with normal glucose tolerance and then develop impaired glucose tolerance, and then fall further down into the diabetes range. Okay? Over time or over whatever effect we might measure.

So as David mentioned, we had recently conducted a clinical trial to evaluate the effects of the investigational medication in Friedreich's ataxia, and we assessed the glucose metabolism in 31 non-diabetic people with FA. We wanted to examine the relationship between various clinical and biochemical markers of the disease and insulin dynamics. Okay? Both the secretion and the resistance to insulin, and what we found is actually, was not terribly surprisingly, we found a distribution of these dots on this curve, and again, this I'll remind you, this is acute insulin response and this is insulin sensitivity index. The dark yellow line, the solid yellow line is actually the average for a large population, about 450 people that were reported in a paper a few years back, and then these other dashed lines show you the distribution of that population, such that the 25th percentile is seen on this dotted line and the fifth percentile is seen on this dotted line, and the 75th percentile, or in other words, the above average folks are on this dotted line and this, or dashed line, and then the dotted line to the top is the 5% that are the highest, most sensitive to insulin and able to produce insulin under circumstances of an IV glucose tolerance test.

We weren't surprised to see that most of the patients with FA had decreased disposition index, consistent with the increase in diabetes, okay, so the actual average was about the 21st percentile, and that was highly significant statistically speaking. But you can see there's quite a bit of variability. There's even one person up here at the very top that seems to be better than most people out in the general population. And that's sort of curious and may be, there may be something special about that individual that might explain why that is. But, we haven't examined that in detail, but we have put people on the curve to sort of see where it is that they fit. Okay?

Then we actually looked at, and as David mentioned, I'm not going to present any of these data, but as David mentioned, there's a fair bit of variability in these measures and in this small population, we saw little changes over time, but they weren't statistically significant, so they weren't worth reporting. However, we did find some measures, some associations within our data that may, perhaps, might be important.

One of the things that we found was that there was a highly significant association between the age of diagnosis of FA and this disposition index. Okay? Such that people who were younger at the age of onset had a lower disposition index. And again, a lower disposition index suggests a higher tendency to develop diabetes. So not surprisingly, people who were affected at a younger age were more inclined to develop diabetes.

We also found that the number of repeats in the gene was closely associated with the disposition index. Now from earlier in the talk, you may recall that I mentioned that there are these repeats, and the more of them that you have, usually the more profound the condition is. Well, we found that there was a clear association between the number of those repeats and the disposition index, so that these are two markers of the condition's severity that are associated with disposition index and so it probably is a meaningful measure, and certainly, we were able to demonstrate its association in a relatively small population.

But, it is a daunting task to have those IV glucose tolerance tests done. We probably wouldn't do them outside the context of a clinical trial. We probably wouldn't recommend going out and having one of these tests done to try to determine your predisposition to diabetes. That's not really very meaningful in isolation. It's probably more meaningful with the passage of time, what, where, what direction things head.

But in summary, the preliminary results of our study suggested that the disposition index is significantly decreased in people with FA, even in the absence of diabetes. And the disposition index was well correlated with age at onset of Friedreich's ataxia and the number of GAA repeats in the frataxin gene. So we do believe that it's a relevant measure.

If we go back to the data, I can, so the question is, "Is it primarily a problem with insulin resistance, or is it a problem with insulin deficiency?" And you can see it's a little bit of both. We have, I need to get some normal data for people who are comparable. We are in the process of pulling that out to try to determine what, you know, how that association looks, but the degree of insulin resistance was much less than I would have anticipated, based upon previous reports in the literature. So these data suggest to me that the beta cell is affected in a meaningful way with Friedreich's ataxia. And, you know, replacing insulin is a simple matter, so I don't, that doesn't concern me so much as the degree of insulin resistance so that the lack of profound insulin resistance I thought, I took as a generally good sign. So, with that I think I'll end. I'm not sure who's next up.