I hear people say frequently, “Oh we never could have predicted these amazing findings.” I slowly raise my hand and say, “Well, actually, the basic science community predicted all of this five, ten, 15 years ago.” Maybe what was most surprising to me as the years went on was so few people actually got that that was a good idea. There's no doubt that there is glucagon being produced in tissues outside the pancreas, and we've been working as a lot to find out if this was also true for people. We've actually spent an awful lot of time trying to solve that particular question. Today, we know that, first of all, the result of all of this work was the identification of GLP-1. So rather than glucagon we found GLP-1 instead. GLP-1 has turned out to be, we can almost say, a miraculous molecule that does a lot of things. And surprisingly, most of these actions are beneficial. Clues to the gut factor date back to the early 20th century. Only analytical advances such as radioimmunoassay and gene sequencing made it possible to solve the incretin mystery. I got involved in incretin and glucagon research many, many years ago. Actually, it was in the early 1970s, where a very close friend and actually mentor and I, he was called Jens Rehfeld, we created the laboratory together. And one of the ambitions of that laboratory was to find out, what are the incretin hormones? After having finished medical school, I joined a group that was interested in incretin research. So basically, my mentor, the head of this department of gastroenterology and endocrinology, Werner Creutzfeldt, he had been working on incretins since 1964, when he was appointed to be a professor of internal medicine. Glucagon was known since 1923, but it was not precisely known where it came from. So eventually people were talking about the pancreas. But that was not a given thing. That was not a straight thing. In 1964, there was just the incretin concept. So people believed there must be gut hormones that stimulate insulin secretion, but none of them had been described. It wouldn't be until 1981 that Alistair Moody, Jens Holst and colleagues in Copenhagen were able to show that the gut factor thought to be glucagon was actually a peptide they called glicentin. And glicentin which is a big molecule where glucagon is located in the middle, but it cannot act like glucagon, that's the important thing. We've suggested that this was proglucagon, the precursor. And the thing is, in those days, people were obsessed by precursor studies because this was new. But then it turned out that it was perhaps not the full story, because proglucagon one had to be a little bit bigger than our glicentin molecule. And it turned out then that that part of the unknown, bigger part of proglucagon was interesting. So the first sign of that came from Joel Habener’s laboratory in Boston, where he was looking at fish and found, what he called, a glucagon-related peptide in that non-glicentin part of the precursor. In the 1980s, I went to a lab in Boston after completing my medical training to learn the new science of molecular biology, and I was told that I should work on the glucagon gene. In 1984, they had just cloned the genes for glucagon, and we saw two new sequences that had never been seen before. And they encoded what we described as glucagon-like peptides, glucagon-like peptide 1, GLP-1, and GLP-2. And my project was to find out, how do cells make these? What do they do? Do they have biological activity? And that's how I was fortunate enough to be at the right place at the right time. We immediately started going for those two sequences and trying to see what we could make of it. We created analyses for these peptide sequences, and we started to map what it looked like in people. And we found that, yes, you could find these stretches in the pancreas where they were together in a big molecule, which we now call the major proglucagon fragment. And they're probably inactive in that fragment. The story could have ended there. But it turned out that also in the gut, where we have glicentin produced, the other half of the precursor was cleaved to cleave out those GLP-1 and GLP-2 sequences. In Joel Habener’s lab in Boston, Daniel Drucker was the first to show that GLP-1 in its truncated form stimulates insulin secretion in rat islet cells. Also in Boston, Svetlana Mojsov and Gordon Weir successfully experimented on truncated GLP-1-induced insulin secretion in the perfused rat pancreas, while Jens Holst and co-workers in Copenhagen were making similar progress in the perfused pig pancreas. So were able to isolate them and measure them. We could measure their secretion. And we found out what the correct structure was. And that was the truncated that we now know is GLP-1 7-36 amide. And were able to demonstrate that this one, in contrast to all the other fragments, could stimulate insulin secretion greatly. We had three papers, several from Boston, the lab I worked in, and one from Copenhagen with Jens Holst, demonstrating that a short form of GLP-1 actually stimulated insulin secretion. and now they share the glory of having discovered GLP-1. Also in 1987, studies from Steve Bloom's lab in London showed that GLP-1 stimulates insulin secretion in humans. However, implications for diabetes A collaboration led to finding the key missing piece. I had a very good friend whom I met in Madrid in 1985, where we presented our stuff for the first time at the a combined EASD/IDF Meeting. And we got very close friends. And so we collaborated from then on. And he was from the Göttingen group in diabetes. If we look back to the discovery of GIP, which was a wonderful agent to stimulate insulin secretion in healthy animals and healthy human subjects, there was the surprise: It works in healthy but not in type 2 diabetes subjects. And that was the point when we started experiments in humans and our expectations were very, very low. We did two experiments: One highly standardised hyperglycaemic clamps, infusion of GIP, infusion of GLP-1 in healthy subjects, and type 2 diabetic subjects, and that showed with GIP almost no effect in type 2 diabetes. But with GLP-1, the insulin responses were like two thirds of what we found in healthy subjects. It was a little bit impaired, but not substantially. And then we did a very, very simple experiment. We just took hyperglycaemic patients with type 2 diabetes who had come to a hospital to initiate insulin treatment because they no longer responded to any oral treatment, no glucose-lowering tablets. And we infused GLP-1 with a dose, I think we were extremely lucky that it was so successful, well, we just tried it. And surprisingly, these patients within hours had a normal blood sugar just by receiving an infusion of GLP-1. The only problem there was you needed to infuse it as a constant drip, intravenously. Otherwise, you would never have seen those effects. When we gave native GLP-1, not we, but as in colleagues in the field, it didn't last for very long. It was rapidly degraded by an enzyme soon discovered to be dipeptyl peptidase 4. And that then was a problem that had to be solved by the pharmaceutical industry. Unexpectedly, the first successful approach for overcoming the rapid degradation of GLP-1 emerged from the venom of the Gila monster. Within it, John Eng and Jean-Pierre Raufmann discovered a GLP-1-resembling peptide they called exendin-4. And amazingly, despite all the efforts in the pharmaceutical industry at Novo Nordisk, at Eli Lilly, at Sanofi, and many other companies, the first approval of a GLP-1 based medicine came from the development of this lizard based protein that we now call exenatide developed by a very small biotechnology company, Amylin Pharmaceuticals in San Diego. Meanwhile, a young scientist at Novo Nordisk explored a different pathway. I first got involved with GLP-1 in the very early 90s. I was a junior scientist at Novo Nordisk at that time, and GLP-1 was one of those ideas that were around for focus, primarily, on diabetes in the beginning. So I led the first team that solved the druggability puzzle based on human GLP-1, and we made the first long-acting human GLP-1 called liraglutide. We did not use any knowledge from exendin-4. We were very focused on staying as close to human GLP-1 in the sequence as possible, because we wanted to avoid antibodies towards the peptide in people. With liraglutide, we only have one change in the human sequence. The first GLP-1 receptor agonist to be approved was exenatide in 2005. Liraglutide followed in 2009. As reflected in the ADA/EASD consensus algorithm from the same year, initially, acceptance was limited due to the subcutaneous route of application, tolerability, and limited clinical data. This is a lifelong therapy, and it's very reasonable to ask, is this safe over a prolonged period of time? And very early on in the story, really at the very end of the 1990s, we saw case reports of people with pancreatitis. Acute pancreatitis is usually diagnosed if two out of three criteria are fulfilled: There is a characteristic very severe abdominal pain. There is an elevation in pancreatic enzymes, like amylase and lipase. And then there is imaging: ultrasound, CAT scan, and the like. And what we did not know at that time is that treatment with GLP-1 receptor agonists will elevate pancreatic enzymes. And the typical adverse events with GLP-1 receptor agonists is nausea, vomiting, diarrhea, abdominal cramps. And I believe many diagnoses of acute pancreatitis were made on that basis. We also saw from studies in animals that there was communication with GLP-1 and the thyroid gland producing a very rare disorder in animals, medullary thyroid cancer. And so colleagues raised the question, is this therapy going to be safe? And based on a limited amount of animal data and some human data, this question arose and became a very important debate in the field. At the EASD Annual Meeting in Lisbon in 2011, the debate came to a head between Peter Butler and Michael Nauck. I got an invitation to participate in the pro and con debate. And the situation was not clear at all. So there were reassuring data, but others that were of concern. And Peter Butler from Los Angeles was asked to present his position that he thought there are major threats associated with GLP-1 receptor agonists. Peter Butler was in this thing pretty much, and he had a lot of projects that made the GLP-1 receptor agonists look bad. He did many of his studies in experimental animals, but he found cells proliferating and everything. So it was scary. And he really, at that time, had the ambition to kill this class of medications. So he said to me, before we had this discussion in Lisbon: “Be sure, three years from now, there will no longer be GLP-1 receptor agonists on the market. But there were data, that I talked about in 2011, that were reassuring. So the information we had was divided. So the mouse and rat thyroid respond very briskly to GLP-1 medicines by increasing the secretion of calcitonin from these C cells. But monkeys, and more importantly, humans don't. And that information was not easily understood by both professional colleagues as well as regulatory authorities. But it took enough exposure with these medicines to show in clinical trials that these concerns were not valid. We learned that humans don't respond in the same way that mice and rats do. We did not see pancreatitis, we did not see thyroid cancer, and we did not see pancreatic cancer. I think we have a very good safety database with all the cardiovascular outcomes trials that lasted for up to five years. So that, and thousands of patients. So I don't think there is much more safety information for many other drug classes like we have it for GLP-1 receptor agonist. As it turned out, cardiovascular outcomes studies also triggered the clinical breakthrough for GLP-1 receptor agonists. I remember very, very well, how much speculation was going on before the first trials reported results. And the first one with GLP-1 receptor agonists was ELIXA with lixisenatide, and it was completely negative. And other trials were already ongoing at that time. And so maybe in the beginning, people had very high expectations, certainly after the publication of ELIXA, those expectations were lowered by very much. And then came the year 2016 with LEADER reporting liraglutide versus placebo. And there it was: a very significant reduction in myocardial infarction, stroke and associated deaths. And, yeah, there was no way of finding this out other than to do those cardiovascular outcomes trials. The companies didn't like the idea, because they are very, very expensive. But I think we have learned so much from them. Moving back in time: In 1996, Svetlana Mosjov co-published on the discovery of GLP-1 receptors in the brain, the same year that Steven Bloom's lab produced similar results. I started to focus on GLP-1 and weight loss already in 1994, 1995. The community of GLP-1 was a very small one. When I went to the European diabetes meeting, at that point in time, right, everyone was interested in insulin and PPAR activators. So people working with GLP-1 was a community of maybe 25 to 50 people. We were always in the smallest room in the back of the conference center. But there was this budding evidence that GLP-1 had effects on the brain, that the GLP-1 receptor was highly expressed in the rodent brain. That was what we knew at the time. And then here in, I'm in Copenhagen, right. But there was a research group led by Ole Madsen, who had an animal model that expressed really high levels of glucagon, GLP-1 and PYY. And those animals starved themselves. So, of course I knew Ole, and I heard about this, right? So that actually inspired me to think. Already back then, there was such a clear evidence that obesity was driving an epidemic of diabetes and many other comorbidities. So I was just thinking, if I have an idea where we can say, we have a new medicine, potentially a new medicine for diabetes with less hypoglycaemia, with a new mechanism, but it's also, at the same time, a medicine for for weight management and weight loss. And why on earth should I not do things at the same time? So this was actually a little bit before the big papers starting to appear around GLP-1 and obesity. I think the first one was in 1996, and then that was in rodents. And then in 1998, there was a paper that this also translated to humans, to people with obesity, to people living with diabetes. And also that there was an appetite reduction. And then we just worked on that over the years, and eventually we showed that liraglutide did have a relevant effect in obesity. It was clear that there was a formidable problem with the side effects, but that eventually was solved by the very slow up-titrations. What it took actually was a brave dose-response study, conducted by the Novo Nordisk company in 2009, published in The Lancet, where liraglutide was dosed up carefully with slow titrations in people up to three milligrams, which was a huge dose in those days. And sure enough, it worked better on food intake and caused a body weight loss of something up to 6, 7 %. And that was important because that gave rise to a series of registration studies. As GLP-1 receptor agonists emerge as first-line therapies for cardiovascular protection in type 2 diabetes and obesity, attention has turned to other cardiometabolic complications that could also potentially be favourably affected by GLP-1. And these new studies, particularly something like the SELECT trial, shows that you can do something about them. You can really do something about it. When we look at the benefits of GLP-1 medicines, for some conditions, weight loss seems to drive the benefits. So what might those be? I think, for arthritis weight loss is tremendously important. We have people with peripheral artery disease where I think weight loss is very important. Certainly for obstructive sleep apnea weight loss is very important. And then inflammation, I think, is a common theme that is important for many of these complications. If you look at people with what we call the low-grade systemic inflammation, which you have in obesity and in type 2 diabetes, and you then give them GLP-1, judging from the circulating markers of this inflammation, you can very much reduce the inflammatory reactions. The other mechanism, which is interesting here, is the effect on the blood vessels. That's why you have the GLP-1 receptors in the capillaries, where you can demonstrate it in a number of experiments that progression of atherosclerosis will be slowed down with GLP-1, that you can reduce inflammation in the vessels and that you have a beneficial action of GLP-1 on all these vessels. And again, this is something that is common to the entire organism, so that if you improve the circulation, you will also have improvements in the brain, and the heart, and the liver, and so forth. And it raises really important questions. If you don't need weight loss to reduce inflammation and to get benefits of GLP-1, then what's the dose response for the benefits of GLP-1 in terms of reducing heart disease or kidney disease? The answer is we we don't know. And then there's a that final thing that I think is really interesting and that is the effect on the reward system. So it turns out that the reward of abuses, whether it's food, or alcohol, or opioids, or whatever it is, the pleasure or the reward of that will be dampened by GLP-1. And that's probably the explanation for these new reports that you have all over the place for abuse situations, and alcohol, and abuse, and all that. And there are more reports coming out all the time. It's really, really promising. Another direction of research is using GLP-1 as a backbone for peptides with different profiles of receptors in different organs. The logical move is PYY, actually. And why do I say that? I say that because, the most effective therapy of type 2 diabetes and obesity is bariatric surgery. How does it work? It works by stimulating the secretion of gut hormones, in particular GLP-1. It also works together with PYY, which is another product of the same cell from the gut. The other logical solution is called oxyntomodulin. Oxyntomodulin is interesting, because it is acting on the glucagon receptor and on the GLP-1 receptor So it is a natural GLP-1/glucagon co-agonist. Glucagon has a pronounced effect on liver metabolism, particularly fat metabolism in the liver. So the trick is to combine GLP-1 and glucagon, and get rid of the glucagon-like effects on glucose by glucagon, but still have the effect on lipid metabolism in the liver. Amylin came into the picture also, and that's interesting, because we already had Pramlintide, but it was not a good pharmaceutical. But the experience was there. So eventually, the Novo Nordisk company again put a fatty acid on the molecule and made it stay alive for a week. And, sure enough, it would start acting on appetite and food intake. And that's what we have now in the form of cagrilintide. The GIP story is more complicated. The old Vincent Marks called it the obesity hormone and was convinced that it was a factor responsible for the development of obesity. And that's the background for why people are now developing GIP antagonists to treat obesity. So we have these GIP/GLP-1 co-agonists also. In my mind, there is no doubt that the majority of the effect is due to the GLP-1 component. It is possible that there is, as reported, an effect on the side effects, so that by inclusion of GIP, you make the side effect profile milder. In retatrutide, there is a combination of GLP-1, GIP, and glucagon. That is a logical combination. Maybe by adding GIP, again, you can get rid of some of the side effects maybe. So, it's so far the most effective weight-losing agent. How will these new medicines be better than the current ones that we have? Will these medicines prove to be as safe or safer than the new ones we have? And what will they be most useful for therapeutically, beyond simply the achievement of weight loss? And we don't know the answers to these questions. With the many compounds that are being studied in clinical trials, my prediction is there will also be a competition which compound is better tolerable than the other ones? Because once you show your drug can reduce body weight by, let's say, 10 or 15 kg, and there are many such drugs, the patients will prefer those where they don't risk this kind of tolerability problems. GLP-1 has turned out to be, we can almost say, a miraculous molecule that does a lot of things. And surprisingly, most of these actions are beneficial and can be used to effectively treat serious conditions like diabetes, obesity and its complications. Maybe in the future even more. My personal take home message after 36 years in pharma is that science is amazing. We can solve a lot of problems. And it's great when we have collaborations between academia and the industry because both of them are great scientific fields. I think also that successfully applied science is always the work of the many. When I look at the story of GLP-1, it's 40 years of careful science. It started with discovery research, looking at first genetics, then peptide pharmacology, then animal physiology and pharmacology. It just took a series of careful iterative studies, brick by brick, to build a scientific foundation. So I think the message of GLP-1 is not get rich quick. So let's not neglect basic science. It really is fundamental for the future success of the human health on this planet. What we now know is that people with obesity have a two to three fold increase in the risk of myocardial infarction, and stroke, and death. So this is a terrible burden that hits a quarter of the adult population, maybe even more. You can really do something about it. You can prevent this from happening. So I think the biggest thing is you can actually do something about it. And that to me is really fantastic.