Beta Cell Pathophysiology and Pharmacotherapy

A leading primary care physician and a world-renowned beta cell researcher discuss how understanding the pathophysiology of beta cells will influence a primary care doctor's choice of drug therapies for their patients. (28:39)  


Dr. Schneider:  Dr. Leahy, thanks for joining us this morning.  In the paradigm we used to think that Type II diabetes was one that really was related to insulin resistance and the beta cell itself was viewed very much as a secondary player in the pathophysiology. My understanding of that is our knowledge of that has actually shifted a little bit.  Why should we care as a primary care community about really what’s going on at the beta cell level?

Dr. Leahy:  I’m just a true believer that our primary care community will be more knowledgeable and be more effective if they better understand the pathogenesis of such a common disease and start to think about contributing factors that may improve or worsen blood glucose values based on that understanding and also be prepared for effective use of all the drugs we have.  I mean the thing that’s so interesting in our world is that we went from having almost no drugs for treating Type II diabetes to an explosion of drugs over the last ten years. In many ways I think the future will be starting to think about using drugs based on presumed pathophysiological benefit as opposed to simple habit -- will I do drug A, then I do drug B, then I do drug C.  And I am convinced that there will be a day and I don’t think it’s light years away where we will be using genetic information to define what drug therapy is potentially going to be most effective in that individual, what other kind of interventions.

I think also from our grooming of the next generation of primary care doctors, their approach to disease, not just diabetes, is very much going to be pathophysiological based and they will have tests available to them that we don’t use today.  And then finally, I think as one just talks to a patient we in some respects have to get away from the concept, “Oh this is a disease of insulin resistance.  There is this genetic imprinting that people carry.  There are environmental factors that negatively impact different aspects of glucose control physiology -- one of them beta cells.  The concept: why is there so much diabetes in developing populations?  Well, you know, there’s a real idea out there that maybe malnutrition early in life has negatively imprinted beta cells so they don’t develop properly.  And that could be a huge public health issue for the future.

So, I think the general issue right now is to prepare our doctor population for the future, but also to make them more effective about how they communicate the pathogenesis to the disease, how they start to think about interventions and how they try and balance all these different drugs that are out there as to which might be most effective for the patient who’s sitting in their office.

Dr. Schneider:  I think that what you’re really talking about is translating knowledge of basic pathophysiology into the hands of clinicians at the bedside so that they can really take into account that understanding in choosing drugs and interventions that will, in fact, work with the best side effect profile, the best efficacy profile and the best safety profile.  I think about this very much in the context of how we’ve evolved with chemotherapy, for example, for cancer and how with some of the newer approaches we’ve gone from, you know, a very blunt instrument with some of the chemotherapeutics to very specific monoclonal antibodies that target cancer cells specifically so there’s less and less collateral damage, if you will, to normal tissue.  I think about how we’ve evolved with insulin therapy and how we’ve moved to analogs and how by understanding pathophysiology and trying to mimic the insulin secretion from the pancreas, some of the newer analogs have profiles that very much are more in line with how our normal pancreas works.

As we then think about really what is at the root of the diabetes discussion here -- understanding sensitivity and understanding beta cell pathophysiology -- we can then take that same leap in the choices that we’re making for patients.  Because at the end of the day we want, again, for patients to have an effective treatment, a treatment that has low rates of hypoglycemia and has rates of weight gain or effect on weight that are pretty much neutral or beneficial.  Because that’s what my patients care about as a primary care doc is for me to treat them effectively, safely, one that doesn’t have many side effects.  And the more we understand basic pathophysiology and choose to mimic that, mimic nature, I think we’ll be able to achieve those outcomes.

So having said all that, I’m wondering if we could transition a little bit into really how some of the newer agents and some of the newer classes of medications really are more pathophysiologic and, hence, maybe are choices that our patients may benefit from.  So I guess I’m kind of opening it up for you to kind of talk about incretins, incretin therapy in the context of what we just went through with a better understanding of pathophysiology.  What do you think?

Dr. Leahy: Okay, so, that’s a great question and I think I would start with that by really taking the theme you just had which is starting to think about using and designing therapies in a disease based on a preferred treatment profile, not just blood glucose lowering, but also from a pathophysiology point of view.  And that’s actually so interesting in the diabetes world because as I said we have all these new classes of diabetes drugs and we’ve had these older classes, some of them for 50 years, the reality is we really didn’t get agents to use that were designed from the get-go to do something important based on what we knew about diabetes.  Most of these drugs until about ten years ago were drugs that pharmaceutical companies just almost accidentally discovered lowered blood sugars in some kind of model and then they evolved into treatments for diabetes -- not insulin obviously, but many of our traditional drugs -- even TZDs quite frankly.  And then we -- ten years ago pharmaceutical companies got very interested in the incretin system -- which I’ll mention in a few minutes -- and started to ask, “I wonder if we have agents that work on the incretin system that would be a useful drug therapy.”  And the short answer is ten years later they are useful and we have agents in the clinic, but then again they sort of started from a design -- wouldn’t it be interesting to see if X happens if we use a drug in that direction -- and that’s our future.

Many of the drugs , or most of the drugs, that will come to us in the future someone decided based on basic science as an important thing to test that ended up being a useful outcome and we’ll get that drug.  So incretins, what are incretins? Incretins are this hugely important physiological system in all of us that helps to control blood sugars after meals.  It is the dominant physiological system in that regard.  So if anyone goes out and stops and has a spectacular lunch or supper with lots of calories and carbohydrates and enjoys their meal, if they don’t happen to have a problem with glucose tolerance and their blood sugars are pretty normal after that meal, they should thank the incretin system.  Because the incretin system is essentially a physiologic connection from the gut and the brain telling islets, “Hey, food is coming, you need to respond appropriately which is to put out more insulin and also turn off glucagon -- two of the key incretin hormones.” [Note: The two main incretin hormones are GIP and GLP-1.] The reason the incretin system is so important is that first of all there are a number of dimensions of that system that have been identified physiologically and are good drug targets.  One of them happens to be something called DPP-4 -- dipeptidyl peptidase-4 -- which is an enzyme which acts to normally inhibit and to turn off the action of these important incretin hormones.  So you can take a drug that’s slows or stops the metabolism so that the incretin system works either better or longer.  The second is one of the key incretin hormones called GLP-1 -- glucagon-like peptide-1.  It turns out that there are drugs now out there that either look like GLP-1 or bind to GLP-1 receptors and, thus, essentially replicate this incretin affect.

The second issue of the incretin system was of interest some years ago and continues to be of interest is because these hormones that are released from the GI tract when we eat and promote these islet affects that I told you, the system is wounded in Type II diabetes -- it’s reasonably defective.  There is clearly defective incretin regulation that occurs somewhere during the process of the disease and we believe that is part of the impaired islet function -- the beta cell dysfunction -- and the failure to turn off glucagon at a meal promoting post-meal hyperglycemia in patients with this disease and so if that’s true, getting agents which actually work on that system predictably would be a positive and they are a positive.  Again, if you look at the drug classes we have, they help to control post-meal blood sugars just in the way it would have been predicted.  And then just the third issue that’s important, and it actually ties into your introductory comments, so we’ve had drugs that promote better insulin secretion for more than 50 years, which are sulfonylureas.  And so you might say, “Well, they’re pretty good. Now why would I want another drug?”  Well, one of the major flaws of sulfonylureas is that they act on a signaling pathway, which is an ion channel right on the beta cell membrane -- right really at the distal stages of insulin secretion and so they bypass all of the important intracellular beta cell biology which makes that cell glucose responsive.  So normally that cell is incredibly precise at regulating insulin secretion to the prevailing glucose concentration.  Good if your sugar is high, even better if your sugar is low because it turns off insulin secretion.  The problem is that sulfonylurea kind of loses that.  And so you can give this drug and promote more insulin secretion than you would want when blood sugars are pretty normal, or frankly, low. And so that’s been one of the downsides of those drugs for years -- a risk of hypoglycemia.  And with the incretin system the concept was well this is an important intracellular system, which is also glucose responsive and maybe these drugs will preserve glucose responsiveness and they do.  And so one of the benefits that is superior to the sulfonylureas is a considerably lower risk of hypoglycemia.  So that’s the reason I think we move in to try to use biology to create new drugs and the newest ones we have are incretins that brings some interesting benefits. 

Dr. Schneider:  So it really does seem like less hypoglycemia is a patient-centered outcome that I’d be interested in, and these drugs are effective in having insulin secreted in a glucose dependent way.  And can you say just a few more words about the difference in weight -- why potentially sulfonylurea might be a little bit different in its effect on weight gain or weight loss as opposed to an incretin therapy.  Can you say a few words about that?

Dr. Leahy: Yeah, this is, I think, one of the most interesting dimensions of this whole story because if it was ten years ago and one was thinking about let’s go and test in a laboratory agents that work on this system and see what we get hoping we’ll get better regulation of islet function in people with Type II diabetes and hopefully it will be a little bit safer in terms of glucose regulation.  So that would be the concept.  Now we skip ahead ten years and again what has been so interesting is of these two incretin hormones -- one GLP-1 we talked about, a second one called GIP, which its doctor name is glucose-dependent insulinotropic polypeptide (a bit of a mouthful) -- but both of these hormones we’ve learned in the last ten years have a biology that far, far is outside simply regulating islet function.  And, in fact, GLP-1 receptors are in most of the tissues of the body and when you give GLP-1 in pharmacological amounts a lot of things happen.  Perhaps one of the most interesting and one of the ones which has helped define a favorable clinical profile of the GLP-1 drugs is there are GLP-1 receptors in the feeding centers of the brain and either in animals or humans when you give GLP-1 therapy there is a satiety effect.  There’s probably an effect also on actually optimizing glucose regulation and peripheral tissue somewhat, i.e., improving insulin sensitivity, and the bottom line is that people lose weight.  Now the fact that sulfonylureas are associated clearly not with weight loss and maybe weight gain is just sort of a different biology because they don’t have an effect on feeding that anybody knows and then this issue of weight gain it’s sort of a generic issue with many diabetes drugs, it’s not simply sulfonylureas.  Part of it we’ve argued for years is that if people are hyperglycemic enough that they spill glucose in the urine if you know give a therapy, any therapy, and insulin would be another clear prototype and drop their blood sugars below renal threshold then they now hang on to more calories that they eat and it’s like they’re eating more even though they’re not.  So that’s one issue and that has been legitimized by research for many years.  I mean that‘s real.

But the second issue is that there is this feeling that drugs that promote insulin secretion are potentially associated with weight gain maybe because of risks of lows, maybe a patient essentially just eats to try and avoid lows especially during the day when they’re physically active.  There is this sort of background behavioral biology people think about with these kinds of drugs that may promote some weight gain.  So the actual mechanism we could argue about it doesn’t really matter.  There is a clear sort of defining profile of drug effects -- GLP-1 receptor agonists for the most part are associated with weight loss, certainly many patients.  The DPP-4 incretin drugs are probably not more associated with weight stability, not drugs we think about with weight gain.  And drugs like sulfonylureas and probably insulin in many patients associated with some modest weight gain.

Dr. Schneider:  Great. So, so far we’ve talked about hypoglycemia, we’ve talked about weight and I’ve wanted to return back to the basic pathophysiology for one second just so we can highlight the current classes that we haven’t discussed yet today.  We’ve talked about sulfonylureas.  We’ve talked about insulin.  We’ve talked about GLP agonists and DPP-4s (inhibitors).  There are other agents on the market and just for us to be clear can you give us a very high level overview about drugs like metformin and the TZDs, acarbose.  Do any of those have any activity directly on the beta cell and how should we be thinking about those in the context of the beta cell pathophysiology?  Just at a high level, can you give us a sense of those other drug classes in the context of the current discussion?

Dr. Leahy:  Yes, so your question was so carefully asked because you used the word directly -- do these other drugs work directly on beta cells and that actually becomes an important part of the question and I’ll tell you why.  So one of the things that we cannot do in the diabetes sort of biology world is try and separate tissues as if they work in isolation.  They clearly don’t.  The glucose homeostasis system is this incredibly interactive system where the brain regulates the liver, and it regulates beta cells and the beta cell carefully controls the hepatic glucose production and peripheral tissues are impacted by adipose sites and vice versa.  It goes on and on.  And the reason this is important is if you actually look at thiazolidinedione, the TZDs, the drugs that I think the average provider would say is a pure insulin sensitizer.  It’s the drugs we have to promote insulin sensitivity, and they don’t have any impact on beta cells.  Well, that’s not so clear for two reasons.  The first is there are lots of studies using TZDs in either pre-diabetes or early diabetes, and you get an improvement in insulin sensitivity, of course.  But, in fact, the scientists try and look so -- these drugs look pretty beneficial at that stage of the disease.  What are they actually doing?  The bottom line is they’re stabilizing beta cell function.  And the concept is that if the beta cell is being driven by this metabolic stress -- maybe some hyperglycemia, other metabolic factors, it over secretes and you get beta cell failure from over secretions. So now I come in with an insulin sensitizer -- I rest, quote, unquote, I rest the beta cell.  That actually allows the beta cell to remain healthy and so you stabilize beta cell function.  So many scientists actually look at these TZD intervention studies in early diabetes or pre-diabetes and conclude this insulin sensitizer stabilizes beta cell function.  So, is that a direct or indirect affect?  Well, clearly indirect affects occur.  We’ve talked about that.  This beta cell rest -- no one’s going to deny that.

The other issue is, however, there may be direct affects.  I am a big believer, in part because it’s what I study in my laboratory, that the signaling pathway which TZDs act on which is called PPAR-gamma that is expressed in beta cells, that’s active in beta cells -- I believe it controls important genes -- so maybe we’ll figure out with time that there’s a direct effect.  But at least TZDs clearly do have either indirect or direct effects on beta cells as well as promoting insulin sensitivity.  So that’s that class of drugs.  The same concept in theory could be thought about with metformin.  Now, metformin’s major physiological effect is to lower hepatic glucose production.  People may know that the drug’s been available outside the U.S. for almost 60 years -- this is not a new drug.  And amazingly we really didn’t have a signaling pathway or direct mechanism that metformin worked on until a few years ago.  There is still a little bit of debate, but it seems to work on a fuel sensing pathway called AMP Kinase which is very active in adipocytes also in peripheral tissues.  But again, analogous to the TZD conversation AMP Kinase is present in beta cells and is actually fairly active.  So, we think about metformin as a liver-specific drug.  We give it to try and control fasting blood sugars, which is a reflection of adipose glucose production.  Whether it would actually have good effects on beta cells over time, directly or indirectly -- don’t know.  Certainly, using metformin in pre-diabetes was clearly legitimized and many primary care doctors are aware of the diabetes prevention program -- the DPP study that used metformin in pre-diabetes and showed a reasonably good effect of prevention of diabetes.

Now, there is a long, long list of other agents that are out there and for the most part they’re not used the same kind of way that the other drugs we’ve talked about are used in the United States.  There’s the alpha-glucosidase inhibitor.  They essentially slow carbohydrate absorption from the gut.  Amazingly, they’re used extensively in Asia and not used much in this country.  We have newer drugs that have come out.  There’s a cholesterol resin, which is now approved.  There’s a drug that we’ve used – a dopamine agonist, which has been used for prolactinomas that’s been recently approved.  There is this sort of long list, but I think the classic drugs we tend to think about are the sulfonylureas, metformin, TZDs and the incretin drugs and, of course, insulin.  And just one sort of comment about insulin -- one of the reasons I think it’s so important that primary doctors have an understanding of the importance of beta cell dysfunction and eventual failure in Type II diabetes is we have to sort of get away from the thinking about insulin therapy in this disease as a last resort and as somewhat a result of a patient’s unwillingness to follow a healthy lifestyle program and understand that some people are programmed for beta cell failure and they’re going to end up with enough failure requiring insulin no matter what.

Dr. Schneider:  I wanted to switch gears here very briefly to the idea of durability.  The ADOPT trial several years ago showed us that the agents that we select do fail over time, and I’m wondering if you can kind of put it into context some of the newer agents that are currently available in the incretin class in the context of the ADOPT trial and really talk just for a few moments about the durability of some of the newer agents in insuring that the beta cell decline is attenuated or is lessened over time.  Can you say a few words about that?

Dr. Leahy: Yeah, so the ADOPT trial was a trial published in The New England Journal maybe three or four years ago, and it’s kind of a modern day UK PDS, United Kingdom Prospective Diabetes Study, which was a study to give patients their first diabetes drug -- Type II diabetes -- and then follow them with time and actually see which drug was superior.  So, the attempt is to figure out if I’m going to start a drug which one should I start.  Now, a really important part of this study is to get into the study people had less than three years duration of diabetes, most of them just fairly newly diagnosed of one year or so duration.  Secondly, they’ve never been on a drug.  Thirdly, they had an average hemoglobin A1c of 7.3 percent.  So, for the providers out there when they think about when drugs get started, you know it’s typically higher than that.  And they were given a sulfonylurea glyburide; they were given metformin, or they were given a TZD and in that study it was rosiglitazone and they were followed really for four years, although there were a few people who went for five years and so the data is kind of interpreted out to five years.  So, what was learned?  Remember people starting with short-duration diabetes with an A1C of 7.3 and still a lot of these people failed on therapy.  So, this concept you can put people on a drug and they’ll be on that drug for ten years is just not so true in diabetes especially in today’s world.  And when you now look among the three drugs -- the glyburide, the sulfonylurea failed fastest, and I think in the diabetes world we’ve sort of decided that sulfonylureas can be effective, but they fail pretty fast.  They’re not the world’s greatest drugs.  Metformin did pretty well though it failed a little bit faster than the TZD, and so the TZD was kind of  interpreted to be the best and that’s how the trial was interpreted and written, though to be fair the differences between metformin and TZDs were modest enough that there’s an editorial in The New England Journal accompanying it and balanced weight gain, risk of congestive heart failure and cost of the TZD versus metformin and sort of decided maybe metformin was better.

So those are the details of the study, but actually how the study gets interpreted because it’s a four or five year study and it’s looking at failure, it’s telling us something about drug durability, i.e., you know, there’s a natural history to the disease and the natural history, we think, is related to beta cell failure.  So if beta cells continue to fail, you will no longer respond to this drug.  That’s the concept.  So, can we do anything to slow or stop this progressive part of the disease, i.e., the English phrase  that’s used -- the doctors’ speak is, “Is durability of therapy” and that indirectly is meant to be can we maintain or prevent a decline in beta cell function?  So if TZDs failed a little bit slower than metformin, it has been said that TZDs have better durability than maybe metformin and sulfonylurea.  The use of durability is a little indirect there and a little complicated and there weren’t so many differences between metformin and TZDs that we know and it’s a five-year study and, I think, people would like a longer study because then the drugs should really separate more if there really is a durability difference.

So now we move into your question which is the incretin drugs and really the question of durability of incretin drugs.  And the theory is they might have better durability because we know that they work on the islet cells to promote better beta cell function, i.e., more insulin secretion at a meal, better alpha cell function, i.e., less glucagon secretion at a meal and there’s this background in non-human systems, i.e., cells in animals, that you actually grow more beta cells that are thought to be healthy beta cells from these drugs.  That there is a GLP-1 physiology and probably GIP that promotes beta cells expanding in their mass -- a biology that is signaling pathways that have been identified to all sorts of things.  And so the thought, and I keep saying thought or the speculation or belief -- these very touchy, feely words -- the thought is that maybe these drugs will have better durability than what we have now.  But if we talk about the ADOPT study, one important finding was -- the study was four years and some people five years and we argue it wasn’t long enough -- so no one’s going to be able to comment on truly durability of effect of any of the incretin drugs until we get out five or more years.  And the second thing is you have to do a study that is carefully, carefully designed to compare the ongoing effectiveness of these drugs against other drugs like the ADOPT study.  So one of our problems now is the manufacturers of some of these drugs -- they’ve got groups of people who are on their therapy who choose to stay on their therapy, presumably because they work, and they follow them for a long time.  People can drop out if it stops working so they just keep reporting on the people who choose to stay in the study and the latest data we have is three years out.  One of the GLP-1 receptor agonist drugs, exenatide, still works in a sizable number of people.  Not the five years that we need, or more, and not in a controlled trial with an active comparator so that you can say after five or six or seven years, well the people who were on the incretin drugs failed less than our other existing drugs, i.e., there seemed to be durability.  So I think the term durability is inappropriately used.  It’s thrown around a lot, saying that we know things we don’t know and, I think, for the incretin drugs there is hope that they might actually have a longer duration of effectiveness than what we’ve had before, but no data, no data that confirms that and we have to use them in the now, thinking about are they working for my patient now as opposed to thinking, well let’s put you in this drug and hopefully five years from now it will still be a good drug for you.

Dr. Schneider: Well, I very much appreciate that.  I’d like to conclude by thanking you.  I think your answers were extremely concise and clear and helpful.  Diabetes is not going away.  This is an epidemic that deserves our full attention. The evolving literature is one that you alluded to many times.  We have numerous findings in the last ten years.  I think the next ten years we’re going to understand more and more about genomics and how we’ll be able to predict where our patients are going to wind up and the primary care community needs to pay attention to these kinds of studies as well as the studies that are the clinical trials as well as the genetic trials -- the clinical trials --that will help us make the best treatment choices for our patients.  Treatment choices that will help those patients achieve their outcomes in a safe way, in a way with few side effects, in a way that has a neutral effect on weight. So, in conclusion, Dr. Leahy, I’d like to thank you for spending time with us today and we appreciate your comments.

Dr. Leahy: So it was a pleasure.  Thank you very much.