In this figure we see the results of an early study in which GLP-1 or saline was infused intravenously to patients with type 2 diabetes. The infusion was given over night and well onto the following day covering a couple of meals. The GLP-1 infusion was capable of restoring fasting glucose concentrations during the night time and nearly normalized the meal-induced glucose excursions after meal ingestion. All in all, GLP-1 has many attractive actions with respect to treating type 2 diabetes. Have a look at this comparison of the diabetic phenotype and the actions of GLP-1. The patients have impaired beta cell function, but GLP-1 can improve beta cell function no matter how you measure it. It is of interest that not only release of insulin is stimulated, but also the biosynthesis of insulin, so that the beta cell continuously can produce more insulin to sustain the release. Also it is important to note that the action really is a potentiation of glucose-induced insulin secretion; this means that insulin secretion cannot be increased unless a certain level of glucose is present, and this in turn means that GLP-1 cannot produce hypoglycemia. In type 2 diabetes, the beta cell mass is decreased, and it is known that GLP-1 can induce beta cell proliferation and reduce beta cell apoptosis in experimental animals and even in human beta cells kept in culture. GLP-1 therefore has a potential for preservation of beta cells, although it remains to be seen to what extent this applies to our patients with diabetes. The patients owe at least part of their high glucose levels to hypersecretion of glucagon, both in the fasting state and after meals, and GLP-1 can effectively reduce glucagon levels. Most patients with type 2 diabetes are obese, and one of the physiological actions of GLP-1 turns out be regulation of appetite, and therefore food intake, and typically during GLP-1 agonist therapy a weight loss will be observed. The patients very often have cardiovascular complications, and GLP-1 has been documented to have a number of beneficial actions on the cardiovascular system. Some of these effects may be secondary to the weight loss but others may be more specific. Protective effects on the heart in response to ischaemia, for instance, improvements of endothelial dysfunction, and maybe improvements of myocardial performance. In clinical studies of GLP-1 receptor agonists, the risk of MACE, Major Adverse Cardiovascular Events, now seems to be significantly reduced. Several cardiovascular end point studies of the cardiovascular effects of the GLP-1 agonists are carried out in these years, and the results of these well powered studies are eagerly awaited, the first may appear in 2016. The problem of turning the GLP-1 molecule into a clinically useful entity is its metabolic instability. Clearly a molecule with a half-life of less than 2 min is not clinically useful. Therefore, molecules with improved stability and duration of action had to be developed. The first GLP-1 receptor agonist to appear on the market was exenatide, a synthetic replica of exendin 4, a molecule with 50 % sequence homology to GLP-1. It was isolated from the saliva of the Gila monster, a large poisonous desert lizard. (Why and how this was discovered is an interesting story which I reviewed in 2006.) Exenatide is a full and equipotent agonist on the GLP-1 receptor, but it is not degraded by DPP-4 and is also otherwise eliminated much more slowly than GLP-1 itself, even if the DPP-4 degradation is blocked. Exenatide therefore can be used clinically, and it is typically injected subcutaneously twice daily, but is also available in a slow release formulation, suitable for once weekly administration. Another approach has been to attach a fatty acid chain to the GLP-1 molecule. Liraglutide is such a molecule. Because of the fatty acid, the molecule will bind to albumin in the blood, and in this way both DPP-4 resistance and a very long half-life of 12 hours are obtained. This means that a high and relatively constant exposure will be obtained throughout the 24 hours of the day, when Liraglutide is injected once daily. Molecules with longer half-lives allowing once weekly administration have also been developed by attaching the GLP-1 sequence to larger molecules, which themselves are eliminated slowly. In general, the GLP-1 receptor agonists, used on top of standard metformin therapy, have good antihyperglycemic effects, with improvements of glycated haemoglobin around 1 % and with 50 to 60 % of the patients reaching therapeutic A1C targets of less than 7 %. In addition, a weight loss of 3-4 kg, but sometimes much more, may be obtained. The side effects are gastrointestinal and usually transient and mild. As mentioned, ongoing cardiovascular endpoint studies will hopefully tell us whether these beneficial effects will translate into lower rates of diabetic complications. Another approach to exploit the antidiabetic actions of GLP-1 is to inhibit the catalytic activity of the enzyme DPP-4. Such inhibitors were originally developed for anti-immune therapies, because DPP-4 is also expressed by certain cells of the immune system, here also known as CD26. However, very specific DPP-4 inhibitors turned out to have little or no effect on the immune system, so these projects were terminated. But when it was discovered that DPP-4 is responsible for the rapid degradation of GLP-1, it was also proposed that inhibitors of the enzyme could be used for diabetes therapy, and it was documented in pig experiments that an available inhibitors could enhance the survival of both endogenous and exogenous GLP-1 and in this way give much larger insulin responses. This led to the development of new, clinically useful DPP-4 inhibitors, first vildagliptin, which was demonstrated in a clinical proof of concept study to lower hemoglobin A1c of diabetes patients to the desired target of 7 % over 52 weeks. Subsequently a rather large number of DPP-4 inhibitors have been developed, and the first to reach the market was sitagliptin in 2006. The antidiabetic activity of the DPP-4 inhibitors is not quite as strong as that of the GLP-1 receptor agonists, but their great advantage is that they are orally available, most of them suitable for once-daily administration, and that they have an extremely benign side effect profile generally with frequencies not different from placebo. On the other hand, they have little effect on body weight. Also the DPP-4 inhibitors are typically used on top of existing metformin therapy, and many are available in fixed combinations with metformin. Both DPP-4 inhibitors and GLP-1 agonists combine well with other antidiabetic agents including insulin, usually with additive effects and this means that extraordinarily good results can be obtained with such regimens without increasing the risk of hypoglycemia or causing weight gain. Therefore, my assumption is that the incretin therapies will remain in the therapeutic armamentarium for several years ahead. The company behind the acylated agonist, liraglutide, tried to optimize the weight losing effect of the compound by increasing the dose. The dose limiting factor has been the gastrointestinal side effects (nausea and vomiting) which are clearly dose related, but with slow uptitration of the dose it has been pssible to more than double the recommended dose, and this is now known to be associated with a further weight loss. High dose liraglutide has now been approved for obesity therapy, providing weight losses of 5-10 % of body weight. Such losses are known to have beneficial metabolic effects and this therapy is also associated with decreasing risk of diabetes development and will often revert prediabetes when present in the obese individuals. Finally some words about diabetes surgery. It really came as a surprise when the surgeon Walter Pories in 1997 reported that a large number of morbidly obese patients that he had operated with gastric bypass, showed an almost immediate remission of their diabetes in addition to their weight loss. It was nevertheless true and his findings have been confirmed in numerous subsequent studies. In a huge meta-analysis of more than 130.000 patients, many of which had diabetes, Buchwald found remission in 84 % of those operated with gastric bypass, but he also reported remission rates of around 50 % in those operated with gastric banding. In this respect, it is of course important how remission is defined and surgeons often call it remission as soon as they don’t have to bother about prescribing antidiabetic medicine to the patients any longer. More strict criteria bring down the remission rates somewhat, but still the majority experiences remission, also in the long run. And long term data are actually available, particularly from the Swedish Obesity Subjects study, the SOS study. Here, close to 5000 patients were either operated by banding or gastric bypass, or were given a standard of care for their morbid obesity. Many of the patients have now been followed for more than 20 years and the major outcome of the trial is a 30 % reduction in the mortality rate in the operated groups. These figures have recently been confirmed in other large studies as well. As expected the incidence of diabetes in the SOS study and therefore also the incidence of diabetic complications were also dramatically reduced in the operated groups. This is also in agreement with other studies . In a study from 2014, the results of surgery were compared with those of optimized medical therapy over an observation period of 3 years. Again the results of surgery were dramatically superior, with near normalization of glycated hemoglobin levels and virtually no need for antidiabetic medication. These remarkable results naturally give rise to questions regarding the mechanisms involved. If we knew what was going on, we might be able to mobilize those mechanisms without surgery and cure diabetes in that way. Let us start with gastric banding. Although the mechanism undoubtedly is complex, a restriction is created that reduces food intake, and as a result the subjects lose weight. Slowly, and in parallel with the weight loss, insulin resistance is removed, and in agreement with diet-induced weight-loss studies, diabetes remission may occur. Without surgery the problem with diet-induced weight loss and attempts to obtain similar losses by life style modifications is that these maneuvers do not last – usually within a year or two every pound that was lost is regained. With banding, the results may be more sustainable. Again the Swedish SOS study has provided long term results demonstrating this. Apart from the improvements in insulin sensitivity brought about by the weight loss, we know of no other endocrine or other mechanisms that are activated by banding. With gastric bypass, the situation is different – here the diabetes resolution is almost immediate, before a noticeable weight loss occurs, and it is actually known that normal weight individuals with diabetes also experience diabetes remission if they are operated with similar techniques, even if they don’t lose weight. Therefore, other mechanisms must be involved. Recent research has pointed to an important role of hypersecretion of gut hormones, including GLP-1 and PYY as important factors. PYY is another interesting gut hormone, which is actually also produced by the L-cells. One of its main actions seems to be inhibition of appetite and food intake. The main factors responsible for the early remission seem to be the following. The figure shows the important features of a gastric bypass operation of the type called Roux-en-Y. This refers to the Y-shaped configuration of the upper digestive system created by the operation. The stomach is divided close to the esophagus, and a small pouch of approximately 30 ml is created from the upper part of the stomach. Next, the small intestine is divided about 75 cm from the pylorus, and the distal end is anastomosed with the small pouch, while the proximal end is anastomosed to a more distal part of jejunum some 1-2 meters from the pouch. This operation was originally designed to create a permanent restriction, by reducing the volume of the stomach, and to create malabsorption by bypassing the stomach and the upper small intestine, but this is actually not what happens. The earliest result of the operation is a dramatic improvement of the insulin sensitivity of the liver, undoubtedly as a consequence of the energy restriction during the first days after the operation. This causes a rapid reduction in liver fat, which explains the improvement. When the individual starts to eat, the food rushes down the esophagus, immediately passes the pouch and enters the so-called alimentary limp of the Y and finally the common limp, where it meets the digestive juices. This presents nutrients to more distal parts of the small intestine, where the L-cell density is higher. This results in a grossly exaggerated secretion of GLP-1 as shown in the figure The response may be 30 fold greater than in non-operated controls exposed to the same meal. The L-cells also hypersecrete PYY, which does not seem to influence glucose metabolism, but which is potently anorexic. Together these two hormones may explain the loss of appetite and thereby the reduction in food intake, observed postoperatively. Plasma glucose concentrations also rise rapidly because of an accelerated absorption of glucose and together with the exaggerated GLP-1 secretion it is no wonder that insulin secretion is also increased. In fact, the insulin secretion pattern may appear completely normal after the operation. The increase is seen already a week after the operation and remains 3 months and 1 year as shown in the figure. In fact it lasts for many years, together with the exaggerated GLP-1 response. It has been proven that it actually is the increase in GLP-1 secretion that is responsible for the augmented insulin secretion, because the augmentation is completely lost if the patients are given an antagonist of the GLP-1 receptor. Because of the increased insulin secretion, deposition of glucose in muscle and fat is also greatly accelerated, and this is the immediate cause of the improved glucose tolerance. With time and as the patient loses weight, ectopic fat is removed also from the muscles and therefore, the peripheral insulin sensitivity is also improved. So the diabetes resolution is caused by an increased secretion of insulin caused by gut hormones and an improved insulin sensitivity, caused by energy restriction and weight loss. There may be other factors involved, and there are many details that I haven’t dealt with, but there is little doubt that these are the essential features. Therefore one the main research challenges in this field for the coming years will be to find out exactly how the exaggerated secretion of the gut hormones is brought about. If we knew that, we might be able to induce a similar hypersecretion without the operation. Naturally attempts are also being made to utilize both of the hormones GLP-1 and PYY for obesity treatment, particularly since it has been shown that they may have synergistic effects on food intake. In other words, the body itself seems to possess the best tools to cure both diabetes and obesity.
