 Thank you. Stephanie is just saying all that because a few years ago I complained about having the slot between lunch, between 11.30 before lunch and she gave me the last slot. So all that is sort of made up. So it's really a pleasure for me to be here. And really, I know this is the last talk, so we don't have to go too dim on the lights. This is fine, but I changed my talk just a few days ago when I spoke to Dr. Coulter when she really asked me to kind of give an update of where we stand now in the new trials. And I hope you indulge me because I decided I would go on a more historical thing, a more historical tour of our landscape, which is again a dangerous thing to do as the last lecture. So I'm going to actually start way back in the beginning for cardiovascular outcomes. And I just want you to think for a minute where you were in 1970. You just think for a second. You know, there were a lot of big things going on at that time. You know, the Beatles broke up in 1970. The computer floppy disk was introduced, so these big events. Richard Nixon was the 37th president of the United States. No editorial commentary. And locally, the clinical facilities and hospital staff were Baylor, my institution, and our institution, and the Methodist Hospital affiliated, no longer together, divorced. And the TMC board approved deeding 5.5 acres of land to the University of Texas as a site of the new medical school in Houston. So big changes here in Houston. So I worked really hard to try and find a picture of a cardiologist who would go on to give a series of successful lectures of women's advocate in health care and a series of successful lectures. And today I find she wears tights to work, or stockings to work. Not really her, but I worked hard to do it. So really in 1970, I think when we think about cardiovascular outcome trials and we think about, in general, the first diabetes outcome trials or the first randomized controlled trial that I can think of was actually published. And that is the University Group Diabetes Program. This was a North American study, which was started in 1961 to 1969 that had a little over 1,000 individuals in five groups. First randomized trial. Placebo, Tobutamide, which was a sulfonioria. Phenformin, which is a precursor to metformin. Insulin and fixed-dose. And insulin and adjusted-dose. Yeah, no, no, this is great, right? And so Dr. Cook said we haven't gotten very far. I had to walk to the library to get this and make a copy because it's not electronic. But if you look at this, this is just the editorial company, the main outcome. And I'll just read it. It says, for some time, differences of opinion have existed as to whether or not the vascular complications of diabetes mellitus can be postponed or prevented by control of the blood sugar in these patients. So in 1970, we're asking the same question we're asking in 2016. So we haven't gotten that far. So this is actually the glucose control in the, in the trial. Now if you look, there are four curves. There's one placebo. There's one Tobutamide, which is a sulfonioria. There's insulin and standard dose in insulin and, I wonder if this works, in the variable dose. Now you remember that I said that there were five arms to the UGDP and Phenformin doesn't make it, right? It doesn't make it to the finish line. It stopped early because of this, because of lactic acidosis and adverse outcomes. So medicine that lowered glucose but had an off-target effect that had to be stopped. So this is a common theme as we go forward that we're learning this lesson that although we can lower glucose, you know, we have to, we may not be doing the right thing. So if we look forward at the outcomes, we can see that again this disconnect that exists. We can lower glucose successfully, but do we get the outcomes we want? And so this is cardiovascular mortality in the study published in 1970. And one can see that right off the bat, that Tobutamide, a first generation sulfonioria, had increased mortality to the other three arms. So an interesting finding again that we lowered glucose but there was this mortality signal. Again explained by different sort of hand waving things, but no one really knew what had happened. Maybe it was a statistical issue, a methodological issue with the study. I want to fast forward now to the mid 90s in the UK PDS study because it says provide some interesting information. As you remember UK PDS was a study of patients who had newly diagnosed diabetes. About 2% of them had heart disease and they were randomized to intensive versus standard glucose control, intensive being around a hemoglobin A1c of 7. In that arm, there in the study there was also a small sub study where obese patients were randomized either to sulfonioria or insulin versus metformin and the outcomes are shown on this slide. So what we see is that metformin was associated with the statistically significant reduction in myocardone farsion as well as in total mortality. Albeit there were a small number of individuals who were enrolled in this study. A little over 350, 340 individuals who really were part of the study. Now this is again interesting because in that metformin group they actually didn't have lower blood glucose in the sulfonioria or insulin. In fact sulfonioria and insulin was somewhat lower than metformin but we saw this cardiovascular benefit and indeed metformin in part related to this remains first-length therapy for people with diabetes in the hopes that it also has cardiovascular risk protection. Now in this historical tour, if you let me just go forward just a little bit more, I want to take you now to the mid-90s. In the mid-90s I was a hot shot medical student and intern and you know at that time if you think back what people were dying of when we were rounding as students, they were dying of AIDS, right? They were dying of AIDS and there was this huge political push that the drugs to treat AIDS were being slowed as far as approval in the FDA. That approval was too slow and people were dying so the FDA decided well we need to fast-track so we can get these drugs to market and so the first drug ever approved on this fast or this rapid fast-track approval happens to be a first in class medicine that treats the underpinnings of diabetes that treats insulin resistance a TZD. Triglitazone, not AIDS, not HIV medicines but really a diabetic medicine and that was Triglitazone. So approved in 1996 and actually what began use in March of 1997 in the U.S. Very successful drug, again first in class that captures 12% of the U.S. market for oral agents for diabetes. Shortly thereafter it's launched in the U.K. but then we start to see trouble. In December 1997 there's 135 cases of liver toxicity as well as success and it's withdrawn from the U.K. markets. At the same time historically there's a study going on the diabetes prevention program which had lifestyle metformin as well as Triglitazone arms and it's withdrawn from the study because there was a death in the study so the NIDD case stops it but it continues here in the U.S. until the March of 2000 when it's pulled off the market and so during this time a very successful drug right TZDs are very good as far as treating insulin resistance and improving HDL, reducing triglycerides, improving glucose. There are 94 patients who had liver failure with 66 deaths in the U.S. So it's a lesson again we have a medicine successful lowering glucose but it has an off-target effect that was unanticipated and although very financially successful for I believe it was GSK. So finally as we think of the landscape this I think was really what kind of changed everything. So we had these lessons and this is a meta-analysis that was published in the Newman Journal in 2007 from Stephen Nissen where he looked at rosy glitazone compiled some data that's been heavily criticized for some methodologic issues but nonetheless this issue that rosy glitazone again very commonly used may increase the risk of myocardium infarction and cardiovascular death. So we had a series of studies which had shown that there were potential adverse effects to some of these medications that can lower glucose. Now around the same time these studies of intensive glucose control come out a chord advance in VADT. These are the studies that said hey we know that high glucose is bad for cardiovascular disease what if we lower it to near normal? Can we reduce CV mortality? Can we reduce cardiovascular outcomes? And these outcomes are shown here on the slide for a chord advance in VADT. In each of the studies there was around a one to one and a half percent difference in hemoglobin A1c but one can see that despite achieving this glucose difference we didn't actually improve cardiovascular outcomes in this group of individuals. Much to our surprise and in fact a chord was stopped prematurely because more people died in the intensive arm than those individuals who were treated with standard therapy. And so lots of reasons why this happened you know these patients had diabetes for a long time they had complications and perhaps it was just too late. And then this is issue of perhaps maybe the medicines that we're using just aren't very good to get them where they need to be. Perhaps the adverse effects related to hypoglycemia to weight gain outweigh any benefits that we see with glucose lowering. So we have that issue with pharmacotherapy and at the same time we have this explosion in the number of people who have diabetes in the U.S. Again if you start when the UDGP study starts in 1960 and you go to 2009 you can see this remarkable increase over time in diabetes and so it's increasing. And then we see that this help burden you know despite how we treated despite our improvements in blood pressure control and lipid lowering therapy that there's this disconnect that exists or this discrepancy in survival between those individuals with and without diabetes this is for men and this is for women just showing that the presence of diabetes shifts survival curves downward and leftward. So despite our improvements in how we treat people this survival is impacted and continues to be part of that. And then finally as we set the landscape we have this explosion of new therapies have become available. These are listed from the recent diabetes care and doesn't include insulin but we can see there's a by guanide, sulfonylureads, the TZDs, DPP4 inhibitors and the different medicines that are available, GLP1 receptor agonists, SGL2 inhibitors and a variety of other medicines that are now available to treat type 2 diabetes. So it's all of this that really influenced the landscape. The unrecognized off-target effects of diabetic medications, the finding that improvement in glycemic control, A1C, is not necessarily concordant with CV risk reduction and the clinical burden of diabetes continues to grow as well as the surge in hyperglycemic therapies. And it's in this setting that the FDA says no more no more we can't do that anymore. So in 2008 the FDA said we're not going to allow this to happen you know all these signals so it says that if you're in industry you have to study these drugs and people who are high risk for cardiovascular disease and it's not good enough just to lower glucose but you have to show that they're safe in these individuals in whom they're going to be used. And this led to this explosion of clinical cardiovascular outcome studies that are now being ongoing and now being completed. These are actually seven studies that have been published in the last year and a half to two years which are part of these CV outcome trials looking at medications not to lower glucose but can these medicines are they safe first it's not a priority and two are they effective at lowering cardiovascular disease and people who have diabetes and who either who have established disease or are high risk for disease. And so we have a few and I'm going to go over the last three in a little more detail but we have them for the DPP4 inhibitors saxaglyptin, alloglyptin, and citaglyptin. We have clinical trials for the GLP1 agonist, lixicinotide, loraglitide, and somaglitide which is not commercially available. And then we have an SGL2 inhibitor empicliflozicin. Now again these are non-inferiority studies predominantly that's how they're powered that's what the FDA said first show is it's as good as and safe compared to placebo. And we can see that for these studies they met their non-inferiority outcome but more recently we see some promise that indeed there may be some benefit associated with these medications to lower CBD relative to placebo. One of our research interests has been in heart failure hospitalization and we can we'll show this in more detail again there may be some hope. But again if we just look at saxaglyptin no one ever thought that this medicine might be associated with greater rates of heart failure until we did the clinical study. So let me turn now to the clinical outcome studies and I'm going to start with the SGL2 inhibitor and empagliflozicin and the empareg study. So I know you're familiar with this but just as a reminder the SGL2 receptors is in the proximal tubule of the kidney and it inhibits the glucose sodium co-transporter causing glucose urea as well as naturesis in the kidney. So this receptors are regulated and people with diabetes inhibits it and so there's loss of urinary glucose excretion as well as some degree of sodium. Now the SGL2 inhibitors are associated in general with the hemoglobin a1c reduction of about 0.7% because of the glucose urea and probably naturesis there's modest effects in blood pressure five and three millimeters their modest weight loss felt to be to the energy and balance or caloric imbalance as you have glucose urea as well as reductions in plasma volume. So what is the empareg study? The empareg study was a study to examine the the long-term effects of empagliflozin versus placebo on cardiovascular morbidity and mortality in 7,020 patients who had type 2 diabetes and who were high risk for cardiovascular disease. These include patients who had a prior MI, coroner disease, stroke, unstable angina or occlusive PAD. They are randomized one to one to one to either 10 milligrams of empa, 25 milligrams of empa or placebo and they were followed for a median of approximately three years. And these are the primary outcomes that were published last year which show that empareg is associated with a significant reduction in the primary outcome of cardiovascular death, non-fatal MI or stroke with a hazard ratio of 0.86. And we can see if we look at the survival curves that indeed this begins to separate early. Now if we look at the individual components the predominant component that drove it in that outcome was cardiovascular death which is shown in here. So for early this study for the first time showed that there was a medicine that was associated with reduced CV death in patients who had diabetes. This is another way of looking at that composite outcome looking at the three-point MACE outcome that we discussed. Again CV death was the one that was significantly reduced. And indeed when we look at the other two outcomes non-fatal MI, coroner infection as well as non-fatal stroke the atherosclerotic outcomes we didn't see any a benefit that there was some heterogeneity that existed within those endpoints. So moving forward looking at something that we've been interested in studying from a personal research area is the effect that it had on heart fader. Again an unexpected finding that there was a 35% reduction in the hazard ratio for heart failure with emporeg relative to placebo. And again interesting because these curves begin to separate very early right very early relative to for example statins or even blood pressure medicines you know the curves just really go apart. And then finally total mortality again mirrored cardiovascular mortality and was reduced with emporeg relative to placebo. So for this study the primary outcome really seemed to be driven by CV mortality with no difference in M iron stroke. The empore and placebo curves began to separate early approximately three months and even sooner for heart failure. There was no heterogeneity between the different doses between 10 and 25 as far as the CV mortality endpoint or the CV outcomes. Mark reduction in heart failure events and in adverse events in general was well tolerated. The biggest adverse event was genital infection was increased with emporeg relative to placebo and this was more common in women. So we were left with this unexpected finding you know why is that and I think the you know why is an sgl2 inhibitor reducing cardiovascular mortality and I think the answer to that no one really knows. There's been several proposed mechanisms you know what could be going on with it related to metabolic changes blood pressure changes myocardial energetics structure function and kidney and I just want to go over that and maybe one by one just look at the different possibilities and at the end of the day I'm going to say I just don't really know what happened. So first let's talk about glycemic control was what the benefit we saw actually related because it controlled blood sugar better. Well it did control blood sugar better you can see here's placebo empotenazine purple empa 25 is in blue and you can see that over the difference there were modest differences in A1C 0.45 and 0.28 so very small differences that you see and we saw in the other studies that glucose controlled where you got differences of one to one and a half it didn't make that much of a difference in mortality. So I think that this difference in glucose control is probably not the reason that we saw the benefit. What about potential effects on blood pressure I told you earlier that it can affect blood pressure so here we see that in general the empa 10 and empa 25 had similar blood pressure control probably related because the empa 10 provides maximum maturesis as a empa 25 but you can see despite the change of five to two millimeters in systolic and diastolic blood pressure that there's no change in heart rate so there's a difference but there's no change but these differences are modest five and two millimeters in a core blood pressure if you go back and think about diabetes those differences were 10 millimeters and we didn't see an outcome. So again I don't think five and two millimeters really explain this difference as well. I think probably the most interesting part of this medicine is perhaps it had some effect on kidney and some is something on volume reclamation and and renal hemodynamics and if you'll spare me I'll just go through this a little bit it so normally we have glucose filtration as well as and there's sodium and glucose absorption that occurs in the proximal tubule. When you have diabetes there's hyper filtration that occurs we know this is the hallmark of diabetic kidney disease is hyper filtration really related to tubular glomerular filtration pressure gradient. So in people who have diabetes they have increased sodium and glucose reabsorption so there is in decreased sodium delivery to the macula densa. So when the when the kidney senses is a decreased sodium delivery what it does is it causes afferent you know it thinks it needs to dilate so it causes afferent arterial vasodilation which in turn will essentially lead to more hyper filtration in people with kidney disease. So what some elegant investigators have shown was in fact in people when you use SGL2 inhibitors that you have increased sodium delivery to the macula densa so you have more sodium going downstream because the co-transporters reduce and in turn there's a feedback that causes afferent arterial constriction which reduces to this tubular glomerular filtration so there's decreased filtration that may occur within the kidney. And so this may explain some of the differences that we can see and in fact in just a publication a few months ago when we looked at EMPA reg again you saw that people who were treated with EMPA reg had improved not only cardiovascular outcomes but improved renal function over time. This is the incidence of incident or worsening nephropathy in EMPA and red and placebo and gray and we can see some marked reductions in the hazard ratio associated with EMPA. So perhaps it's something related to cardiorenal that we're seeing related to heart failure that in turn may have something to do with cardiovascular mortality. I'm sure that cardiorenal was discussed in part this morning with heart failure with preservative where it may play a bigger role. And finally there's the issue with heart failure that maybe we're just seeing some fancy diuretic that people are losing volume in patients who may predispose to heart failure they may have stiffer ventricles and indeed if we cause a little more diuresis that we'll see improved heart failure outcomes. I think that may be helpful for heart failure but I don't think that explains cardiovascular mortality because the diuretics in really that we use for treatment of heart failure as we heard in the previous lecture really aren't associated with improved mortality they're really associated with improved symptoms. This is just showing that you get hemoconcentration which we think may be in due part related to diuresis or hemoconcentration. So that's possible. So I'll say that we really don't know why we saw that benefit but despite that last week, last week, December 2nd, I'm really behind in my Christmas shopping. It's moving really quick. I think that was last week. The FDA made a big big change. The you for the first time the FDA approved a medication a new indication for embolgophosis and to not reduce glucose but to reduce the risk of cardiovascular death in adult patients with type 2 diabetes mellitus and CV disease which is a big step for the FDA and I think we still have to I mean just sort of hot off the presses and how we deal with this in the future I don't know as far as treatment algorithms. So I want to finish just with the last two studies as far as GLP1 agonist. Again I know this audience from there with them but just this point of review the GLP1 is secreted in the intestine when we ingest food. GLP1 causes insulin secretion decreases satiety it reduces appetite it decreases gastric motility increase decreases glucagon production all felt to have the glycemic effects that we know are part of GLP1 agonist. We know that DPP4 inhibitors inhibit the enzyme that break GLP1 and that's how they work. So here's GLP1 as far as their metabolic effects but there's a lot of literature suggesting that perhaps GLP1s have a potential cardiovascular effects. This is actually a publication that we published five years ago looking at just summarizing some of the effects some of the studies that were available both animal models as well as some clinical studies looking at the effects that GLP1 had on blood pressure, modest effects on blood pressure as well as weight, some human studies on endothelial dysfunction, heart failure, MI, and ischemic reperfusion. And when we published a study we thought it was a good paper, worked hard on it. I actually probably never thought it was true but when we look at it I think we see some promise. In fact a study that really showed that there may be some benefit beyond just glucose control is the leader study which was a study of loraglotide in cardiovascular outcomes in patients with type 2 diabetes. Took a little over 9,000 individuals with type 2 diabetes who were high cv risk and randomized them to 1.8 milligrams of loraglotide or victose versus placebo. Included people with renal dysfunction which was a unique aspect to this study because they're often excluded and treatment options are limited in that group. Again the primary outcome was similar to all the studies. It was a MACE outcome followed for 3.8 years. The average age was 64, 35 percent women and 82 percent had established cardiovascular disease. And these are the outcomes. Here on the left is the primary outcome, the MACE outcome, and we can see that there was a 13 percent reduction in the primary MACE outcome with loraglotide compared to placebo. If we look over these outcomes again looking at death from cardiovascular causes the hazard ratio was 0.78 against statistically significant favoring loraglotide. Nonfatal myocardial infarction trended in the same direction but was not significant and there was no difference in nonfatal stroke as far as their outcomes. So other observations in leader is that A1C when we look again looking for explanations or differences between the groups hemoglobin A1C was reduced modestly in the group. There wasn't a huge difference when you compared them at the end of the study. There was modest weight loss associated with loraglotide as well as minor changes in blood pressure albeit statistically significant. Again there's a signal that there's lower rates of nephropathy that occurred with loraglotide compared to placebo in the study. For those of you who use GLP1 agonist you know that the most common adverse effect is nausea, vomiting, and some of the GI effects and that that was increased in the loraglotide group compared to placebo. And I think what probably doesn't make the sound bite but we do have to keep an eye on it is there was this numerically greater pancreatic cancer that was seen with loraglotide versus placebo. It was 13 versus 5 a p-value of 0.6 and I think we'll just need to keep an eye on this signal there's always been a concern about the effects of GLP1 on pancreatitis, pancreatic cancer, medullary cancer. This is the first time I think that I can see from a randomized trial where there does seem to be this imbalance so I think we'll have to look at this in the future as we move forward and I think it echoes what we heard about at the beginning that with these medications we really have to be concerned about off-target effects. No one should really die today from a treatment of type 2 diabetes from a medicine related to it when we have so many options so we need just to keep an eye on that. And then finally the last study that I want to mention is published last month in the New England Journal which was a study of semaglotide unlike loraglotide the semaglotide is is currently not approved it's similar to xenotide or bidurium which is a once weekly injection so it has a long half-life. This study was smaller around 3,300 patients with diabetes and high cv risk and again uses different things one one to one one 0.5 milligrams or one milligrams once weekly of semaglotide or placebo so they were matched. Again the primary outcome was cardiovascular death non-fatal amyro non-fatal stroke and a shorter study this was 2.1 years again it was primarily a non-inferiority study but when we look at it and we look again at the primary outcome we can see with this long-acting glp1 agonist that there was improvement in the the outcome was lower in the semaglotide arm relative to placebo again a novel finding in these studies non-fatal myocardial infarction trended to be lower in this study non-fatal stroke was lower compared to placebo and death and cardiovascular causes were similar. Again if we look at the primary outcome again a benefit in this most recent study there were differences in the two arms as far as a1c the semaglotide 10 there actually was no difference between the two doses as far as cardiovascular outcomes but there were some differences that are shown on this on this slide related to a1c weight blood pressure as well as heart rate so we do see some differences related to the medications it did affect hemoglobin a1c but the cardiovascular outcomes were similar in both studies so this is just a reminder again that again when we read these studies we have to go beyond just the primary outcome and we really have to look at adverse effects and there was an unexpected signal that was seen with retinopathy with this medication now we know that there are lots of tests that occur you know there are lots of outcomes that are looked at and so whether this is just a play of chance or whether it's real it's unclear but nonetheless when we look at the incidence per 100 person years semaglotide had a greater incidence of retinopathy complications relative to placebo a hazard ratio 1.76 so again no free lunch and we need to keep an eye on these outcomes so i think i'm that's going to stop here and really thank you for your attention that you know we talk about the landscape of diabetes and cardiovascular disease it has been shaped by multiple factors including the continued and growing burden of diabetes and cvd the realization of potential off-target hazards of anti-hyperglycemic medications beginning in the 1970s and continuing to last month the notion that we have a large number of treatment options available today many choices both between classes as well as within classes and that there has been some promise really what in my field of what we do research this is really remarkable that for once we're actually seeing drugs that have promised as far as reducing cvd which is still the number one reason why people with diabetes die but we definitely need future studies to define optimal treatment strategies and algorithms and decide what medicines we use first and when and in what combination so thank you