What are the effects of using a GLP-1 drug when on the Atkins diet and a low-fat diet? Who are the best candidates for using a GLP-1 drug? Can GLP-1 drugs reduce apoptosis and increase betacell regeneration? Our publisher, Steve Freed, asks these questions and much more of Dr. David D’Alessio, Associate Professor of Medicine, Division of Endocrinology, University of Cincinnati, and the Albert Vontz chair in Diabetes Research….
Steve Freed: At the ADA, you presented information on facts and fictions on beta cell preservation in type 2, and after the presentation there was a question from a researcher who argued the point that glucose toxicity was not responsible, or the major responsibility, for beta cell failure.
Dr. D’Alessio: I think what he was saying was that glucose toxicity wasn’t the major cause of beta cell failure.
SF: I was always under the impression that normal blood sugars improve beta cell function dramatically. What has your research shown?
Dr. D’Alessio: Well, I don’t do research in that area. But what I did was I read a lot of papers to get ready for that talk. I think the misunderstanding was he thought I was saying that glucose toxicity was the whole story, and it’s not the whole story, but it is the one thing where we have the best evidence in humans. I listed a bunch of things that could go wrong with the beta cell, from increased reactive oxygen species as a consequence of high flux of fuels, or unfolded protein responses because of a strong drive to make lots of insulin, inflammation – all those things could be part and parcel of beta cell failure in any one person, and they’re not independent of glucose toxicity.
The point I wanted to make about glucose toxicity is that in humans it’s something you can test. You can take a human with a high blood sugar, you can lower the blood sugar, and you can see if their beta cell function is better. That’s been done for 40 years now. The first demonstration I could find of that phenomenon was 1976. So I think lowering blood sugar provides an improvement in beta cell function in a lot of diabetic patients.
There are a lot of questions about that. Does it work in everybody? I don’t know. Does it work in people with really advanced diabetes who have had diabetes for a long time? Don’t know that either. If you do it, how long does it last? It seems to last weeks to months. If you bring the blood sugar right back up, does it all go away? I don’t know that answer. But I think the point with the glucose toxicity thing, is simply as a demonstration that there is some recoverable beta cell function in people with type 2 diabetes.
SF: If you have to define glucose toxicity, would you say that normal blood sugars would be the best way to prevent glucose toxicity? And how would you define a normal blood sugar? According to the ADA, a postprandial of 145 is okay.
Dr. D’Alessio: Blood glucose levels in humans are a continuum. I mean, we say normal fasting blood sugars under 100, but there’s a big difference between 99 and 82, at least numerically, and there is some evidence to say that if you have people with 82, they’re metabolically more fit than 99, although we don’t define 99 as disease, mostly because it’s hard to find reliable, measurable consequences of that. It’s like, how low should your blood pressure be? Well, as low as it can be without you falling over. How low should your blood sugar be? As low as you can regulate it.
But to say that the ADA has set their postprandial cutoff at 145 because they know that above that you get glucose toxicity, I think, stretches the limits of knowledge. We don’t know that. And in fact, we happen to be doing a study where we’re lowering blood sugar and measuring insulin secretion, and I showed a slide on that, and it works beautifully. But we don’t have enough patients enrolled for me to say it works better in the guys that start better, or it works better in the guys that start worse.
And I couldn’t find a number in the literature to which I could ascribe glucose toxicity. Most of the studies that have been done take people with pretty abnormal blood sugars — 200 or above — lower them down under 150, on average, and see the effect. I mean, I think in the clinic, what we find is that you struggle to get somebody under control. Their A1C is 9.5; their fasting blood sugar is 240. You put them on insulin and/or other drugs and you start to walk down the glucose by titrating the dose, and it’s really slow going. Then, at some point, the blood sugars get normal and everything really smoothes out for a while.
That’s the clinical anecdote that we see over and over again. Is that a function of getting to a place where you’ve eliminated glucose toxicity? I mean, that’s an easy connect, that there is reversible beta cell dysfunction and you can get to a place where all of a sudden the control seems to fall into place easily. Are those one in the same? I don’t know. I’m always cautious to make those sorts of easy connections, just because it seems right, without any proof. But it makes sense to me.
SF: Well, obviously you want to get it as low as possible without the hypos.
Dr. D’Alessio: No question. That’s exactly it. Hypos are very limiting. Luckily, in the typical type 2 patients that we see, they’re not nearly as susceptible to hypoglycemias as type 1s.
SF: And I noticed on your website it mentions that you had done studies in animals and humans that have shown GLP-1 is secreted in response to ingested nutrients, especially dietary fat, and that it appears to inhibit food intake. But haven’t we seen that the low-fat diets, over time, make people more obese and it’s really the excess carbohydrates that, stored as fat, that’s the real culprit?
Dr. D’Alessio: I think that’s an oversimplification. My joke is that most Americans are on a high-fat, high-carbohydrate, high-protein diet, and that that seems to be associated with steady increases in average body weights and rates of obesity. I think, again, the dietary literature is pretty scattered, and we’ve done some of those studies ourselves. We did a study six or eight years ago comparing the Atkins diet with the standard, at the time, AHA Step 1 diet with 30 percent of calories as fat.
We found that in four and six months’ time that the Atkins diet outperformed the standard low-fat diet, and subsequent groups showed the same thing, and they also showed that if you followed them up for a year that everything came back to normal, that the effect of a low-carbohydrate, high-fat diet was pretty short-lived. And our conclusion from that was that when we recommend these low-fat diets, most of the distribution of calories is kind of similar. I mean, the average American still probably eats 35 or 40 percent of their calories as fat, but they eat more carbohydrate.
When you put them on a low carbohydrate diet, you’ve shifted everything, so that their whole set of food choices has to change. If they’re going to follow the diet, you know, they only shop in two aisles of the supermarket instead of all 13. And I think in the short term that big shift in food choices actually helps people limit what they eat and they lose weight. When we would do the end-of-study interviews with our patients, they would all say things like, “Man, I’m dying for a piece of bread.” So it was like they were able to hold on. We had this list of how to follow a low-carb diet and they were hanging in there until the end of the study, but they were starting to have the kind of drives that we think cause recidivism in diets.
So again, to say simply that carbohydrate is the culprit, I think is really an oversimplification. I think that just like type 2 diabetes won’t be one set of things that gives you a fixed beta cell defect and insulin resistance, that different people will get there through different combinations of beta cell failure and insulin resistance, and the insulin resistance will be spread out evenly through tissues. I think with diet, the same thing. I think there are some people in whom overconsumption of calories is driven by carbohydrates, and I think in others it will be people whose overconsumption is driven by fats.
But the reality is most of us don’t eat fats and carbohydrates. We eat meals where the two are mixed together, and the overeating tends to be of foods and not macronutrients. I think that what drives people to eat right now is that we have easy access to very tasty, calorically dense foods and that’s what drives them. I think the hedonic part of this is a big thing, and I think that’s independent of whether it’s carb or fat.
SF: And have you seen, as far as GLP-1 goes, that people that have had diabetes for 10 or 15 years, fairly elevated blood sugars, don’t secrete GLP-1 as effectively as those recently diagnosed? Which leads me to the question: who are the best candidates for the GLP-1 therapy?
Dr. D’Alessio: I think much has been made about GLP-1 deficiency in type 2 diabetes, and my view is that that’s a very spotty literature, that you can find some papers that show that and others that don’t. I think the ones that tend to stick in people’s minds are the ones where there looks to be a GLP-1 deficiency. I mean, that seems to make sense — maybe that contributes to diabetes. But I think even when you look at the papers where there is a deficit in GLP-1 secretion, it’s not a very big deficit. It’s not like other deficiencies. It’s not like the insulin deficiency of type 1 diabetes where it’s a difference between some and none. You’re talking about 20 percent decreases. And GLP-1 is a potent hormone, but I don’t think 20 percent makes a big difference, and beyond that, if you do studies you see some high responders and low responders in the diabetic groups.
So my feeling is that the deficiency has been overstated. It may be there but it’s subtle and it’s been overstated. That being said, I can tell you how we use the GLP-1 drugs in our clinic, and what sets them apart, I think, is the fact that some people are going to get pretty good weight loss with them, and that’s a very welcome thing to a lot of people with diabetes. And so we tend to use them as add-on agents in people who are gaining a lot of weight with conventional therapy. So everybody in our clinic starts on metformin unless they can’t tolerate it.
If we go to second-line agents — and in our clinic that would be pioglitazone/sulfonylurea — and in the course of that addition, if people’s blood sugar goes down but they gain a lot of weight, they’re sort of marking themselves as people who are going to do that with other treatments, like insulin, and we would add or go to or substitute a GLP-1 agonist in that case. My personal practice is pretty conservative and so I use new drugs only where I have really clear niche, but that is the niche we see because I think they are as effective as the other agents on the market in lowering blood sugar.
We have less experience with them, but the fact that they’ll not consistently make people gain weight and sometimes have them lose weight I think is a real edge, and so that’s for us, where we’re using them. Even in patients with insulin treatment, even though that’s an off-label use right now, I think that can be an effective combination — basal insulin treatment and then prandial or daytime GLP-1 agonist treatment.
SF: Would you ever consider using the GLP-1 for someone who’s on the verge of being diagnosed with diabetes with an A1C of 6.2, 6.4, something like that, to help them lose weight and improve their glucose?
Dr. D’Alessio: I know that’s being done off-label. I know that’s being done in a lot of places. I know there have been, for years, weight loss clinics that prescribe humongous doses of exenatide, for example, not only off-label but outside the usual recommended doses. We tend not to do that. I think there is pretty good evidence that, in somebody that has pre-diabetes, you can delay the progression with lifestyle, with metformin, with TZD. The TZD seems to be the most potent. On the other hand, it also seems to be the one that promotes weight. But I think there are three well-established, good choices. At least one of those, metformin, is also inexpensive. So that’s where we’re going now.
Again, I am open to the possibility that GLP-1s not only would do this but would be especially good at it, but when you have some stuff where there’s really good trial data and other stuff where you’re just sort of guessing, it’s always hard for me to choose the one that makes sense but it hasn’t been studied, and so that’s why I don’t do that. But I agree with you. I’ve thought for a while that that might be a really great niche for these drugs, but I think somebody — either a company that makes a GLP-1 agonist or a federally sponsored, multi-center group — is going to have to show that, because we know how to do those studies.
SF: From your knowledge, what is the effect on beta cells as far as using the GLP-1s, in reducing apoptosis, and the ability to stimulate regeneration of beta cells? I know they’ve done it in animal studies, so what would you conclude from that?
Dr. D’Alessio: As I said in that talk, one of the things that I think has really driven this whole field of beta cell regeneration and interference with apoptosis and trying to preserve beta cell mass is the data that comes from animal studies with the GLP-1 drugs, and those are really dramatic. In a number of different diabetes models in rats and mice, you can show these really prominent effects. I think, as people have thought this through, the effects are much harder to claim in humans, and there are a couple of reasons for that. It looks now like rats and mice really expand and contract their beta cells to meet environmental challenges.
For example, pregnant rats get these huge hypertrophic islets and then after delivery, when they don’t need a lot of insulin, the islets contract. I think the beta cells are much more plastic in rats and mice; they turn over faster. And so I think that’s one of the ways that they adapt to new challenges is actually by growing and dying. So I think things like proliferation and apoptosis are really important in rodents. I think it seems that it’s much less so in humans. Human beta cells turn over a lot less frequently. We don’t know how much but the estimates that have been published are up to three or four years for the turnover of a beta cell, and some people still maintain that unless you have a real toxic stress you don’t really regenerate beta cells; they’re pretty static. So I think the system is different in humans.
Moreover, although it’s plausible that really long-term treatment with a GLP-1 agonist could slow the decline in beta cell function that we see with other treatments, there is no trial data to support that. Now, I think that there are trials planned that will compare, say, GLP-1 agonist with a TZD, with other drugs, over time, and measure glucose control and some surrogate for beta cell function, and so we’ll be able to see how that plays out. But so far we don’t have good evidence for that. It’s kind of a mixed bag.
On the one hand, the GLP-1s have really driven this area of research, and I think that’s been a really valuable contribution, kind of an indirect contribution to that drug class. On the other hand, we don’t yet have a definitive answer, and I think that’s another thing. It’s like the GLP deficiency in diabetes. It’s easy to overstate because it makes a good story, and we all operate in this sort of sea of unknowing all the time, and so a good story is really appealing. But when you look carefully at what’s known, I think that’s an easy thing to overstate.
Dr. David D’Alessio is an Associate Professor of Medicine, in the Division of Endocrinology, and holds the Albert Vontz chair in Diabetes Research. He joined the faculty at University of Cincinnati in 1999. Dr. D’Alessio has received national recognition for research on gastrointestinal hormones and the regulation of insulin secretion. He has published widely in this area, as well as in related aspects of energy metabolism and nutrition. Dr. D’Alessio is ABIM certified in Internal Medicine and Endocrinology and attends the Diabetes and Endocrine clinics at University Hospital and the Cincinnati VA hospital. He is a member of the Endocrine Society, American Diabetes Association, American Heart Association, and serves on the national council for the American Federation for Medical Research.
The primary focus of Dr. D’Alessio’s research is the regulation of insulin secretion and glucose tolerance in type 2 diabetes, and specifically the influence of intestinal hormones on these processes. In addition, he has additional projects to investigate the actions of GI peptides to control food intake and body weight. Central themes that are common to all of this research is the interaction of ingested nutrients with endocrine signaling systems, and the interface between hormones and the nervous system. The overall goal of this work is to better understand the physiologic mechanisms underlying nutrient metabolism and apply this knowledge to the treatment of diabetes and obesity.
Dr. D’Alessio joined the faculty at the University of Cincinnati in 1999. He has published widely in gastrointestinal hormones and the regulation of insulin secretion as well as in related aspects of energy metabolism and nutrition. He is ABIM certified in Internal Medicine and Endocrinology and serves as an attending in the Diabetes and Endocrine subspecialty clinics at the Cincinnati VA Hospital. He is a member of the Endocrine Society, the American Diabetes Association, and the American Federation for Medical Research. Dr. D’Alessio is also a faculty member in the Pathobiology and Neuroscience programs at the University of Cincinnati.
Copyright © 2011 Diabetes In Control, Inc.,