The GLP-1 family tree — from Exenatide to Retatrutide

Start with the receptor. GLP-1 — glucagon-like peptide-1 — is a hormone your gut releases in response to food. It binds to GLP-1 receptors in the pancreas, the brain, the stomach, and elsewhere, producing a coordinated set of effects: insulin secretion goes up, glucagon secretion goes down, gastric emptying slows, and the brain registers satiety. It's part of the incretin system — the system that matches insulin output to food intake — and it was characterized by researchers in the 1980s and 1990s before anyone thought seriously about drugging it. The problem with the natural molecule was that it was nearly useless as a therapy: GLP-1 is degraded by an enzyme called DPP-4 within two to three minutes of entering circulation. By the time you could inject enough of it to matter, it would be gone.

The first solution came not from a laboratory synthesis but from a desert lizard.

Exenatide was approved in 2005 under the brand name Byetta. It was derived from exendin-4, a peptide found in gila monster saliva that binds the GLP-1 receptor but is structurally resistant to DPP-4 degradation. Where human GLP-1 lasts minutes, exenatide lasts hours. That was enough. In clinical use, Byetta required twice-daily injections — a significant burden for patients — but it worked. Blood sugar fell. And in the trials, something unexpected kept appearing in the data: patients were losing weight. Not as a side effect exactly, but as a coherent biological consequence of what GLP-1 receptor activation does in the hypothalamus. The satiety signal was real.

Exenatide's second iteration, Bydureon, extended the dosing interval to once weekly by encapsulating the molecule in microspheres that released it slowly. This was a pharmacokinetics improvement rather than a pharmacology innovation — the molecule was the same, the receptor target was the same — but it mattered enormously to patients and prescribers. Weekly injections versus twice-daily injections is not a minor distinction for people managing a chronic condition. Bydureon was approved in 2012 and extended the commercial life of the exenatide franchise while the next generation was being built.

Liraglutide arrived in 2010 under the brand name Victoza, approved for type 2 diabetes. It was the first human GLP-1 analog — a molecule built on the structure of actual human GLP-1 rather than exendin-4, but modified with a fatty acid chain that bound to albumin in the blood and slowed degradation. The result was a half-life long enough to support once-daily dosing. Liraglutide's weight loss profile attracted enough attention that it was later approved at a higher dose specifically for chronic weight management, under the brand name Saxenda, in 2014. This was the first moment the pharmacological class publicly declared that weight was a primary target rather than a secondary benefit.

Dulaglutide entered the market in 2014 as Trulicity, a once-weekly GLP-1 agonist attached to an Fc fragment that extended its half-life further. Dulaglutide was designed with patient adherence in mind: a pre-filled, autoinjector pen with a needle that patients never saw. It captured a large share of the diabetes market partly on pharmacology but substantially on the engineering of the delivery device. This was a period when the therapeutic space was competitive enough that patient experience — not just efficacy — was driving prescribing decisions. Trulicity became one of the best-selling diabetes drugs globally.

Then semaglutide changed the scale of everything.

Semaglutide, developed by Novo Nordisk, is a GLP-1 analog with structural modifications that extend its half-life to approximately one week and dramatically increase its binding affinity to both the GLP-1 receptor and to albumin. It was approved for type 2 diabetes as Ozempic in 2017, and for chronic weight management as Wegovy in 2021. The STEP clinical trial program — a series of obesity trials — demonstrated mean weight loss of approximately 15% of body weight over 68 weeks at the highest approved dose. Nothing in the history of non-surgical weight management had achieved that. Semaglutide also came as an oral tablet, Rybelsus, approved in 2019 — the first oral GLP-1 agonist, though with substantially lower bioavailability than the injectable form and requiring administration under highly specific conditions to work at all.

The cultural rupture happened in 2022 and 2023, when the gap between the number of people who wanted Ozempic and the supply of Ozempic became enormous and visible. But by that point, the next generation was already in trials.

Tirzepatide, approved as Mounjaro for type 2 diabetes in 2022 and as Zepbound for obesity in 2023, represented the first step beyond single-receptor agonism. It is a dual agonist: a single peptide that activates both the GLP-1 receptor and the GIP receptor simultaneously. GIP — glucose-dependent insulinotropic polypeptide — is the other major incretin hormone, released by the small intestine in response to fat and carbohydrate. GIP receptor activation contributes to insulin secretion, enhances GLP-1's appetite effects through complementary central signaling, and may influence adipose tissue metabolism directly. The SURMOUNT trials showed mean weight loss of approximately 20-22% of body weight over 72 weeks, meaningfully exceeding semaglutide. The dual mechanism was doing something that single GLP-1 agonism couldn't match.

The logic of the tree branches from here into multiple directions simultaneously.

One branch adds glucagon. Glucagon is GLP-1's ancient sibling — they're both cleaved from the same precursor peptide, proglucagon — but where GLP-1 lowers blood sugar, glucagon raises it. This made glucagon receptor agonism look dangerous from a metabolic standpoint for decades. The reconsideration came when researchers realized that glucagon receptor activation also increases thermogenesis: it raises energy expenditure by acting on brown adipose tissue, and it promotes hepatic fat oxidation. At the right dose ratio — enough glucagon to drive thermogenesis and liver fat clearance, but balanced against GLP-1's insulin-stimulating effect — the combination might burn fat more aggressively than GLP-1 alone.

Retatrutide, Eli Lilly's triple agonist currently in Phase III trials, tests this hypothesis fully. It activates GLP-1, GIP, and glucagon receptors simultaneously. Phase II results published in the New England Journal of Medicine in 2023 showed mean body weight reduction of approximately 24% over 48 weeks at the highest dose — exceeding tirzepatide's Phase III data in both magnitude and speed, with the caveat that Phase II and Phase III data aren't directly comparable. Phase III results are still emerging, and retatrutide is not yet approved anywhere. But the direction of travel is clear.

Adjacent branches in the tree take different angles on the same terrain. Survodutide, Boehringer Ingelheim's dual GLP-1/glucagon agonist currently in trials, is being developed primarily for MASH — metabolic dysfunction-associated steatohepatitis — where the glucagon component's effect on liver fat appears particularly relevant. Mazdutide, Innovent Biologics' GLP-1/glucagon dual agonist, has completed Phase III trials in China with robust weight loss results and represents a different geographic and economic vector on the same pharmacology. Cagrilintide, an amylin analog being developed by Novo Nordisk, acts on a completely different receptor — the amylin receptor, involved in satiety signaling from a distinct pathway — and is being studied in combination with semaglutide as cagrisema, on the theory that hitting multiple satiety pathways simultaneously produces greater weight loss than hitting one.

The tree also extends toward longer duration. Orforglipron and danuglipron are oral small-molecule GLP-1 receptor agonists — not peptides, but small synthetic molecules that activate the same receptor. The advantage is that they don't require injection and have more conventional bioavailability than Rybelsus. The disadvantage is that small-molecule receptor agonists can behave differently at the receptor than the natural peptide, and the efficacy and tolerability data are still maturing.

What does this tree reveal about metabolic medicine?

It reveals that the GLP-1 receptor was one lever in a system that has many, and that drugging one lever well — semaglutide — was enough to reshape an entire cultural conversation about weight. It also reveals that the system has more levers than one. GIP adds something. Glucagon adds something. Amylin adds something. The question being worked out across a dozen active Phase II and Phase III programs right now is which combination of levers, hit at which ratio, produces the best outcome with the most acceptable side-effect profile for the most people.

It also reveals how pharmaceutical generation works: each drug improves on the last not by revolution but by targeted iteration. Exenatide solved the degradation problem. Liraglutide solved the human analog problem. Semaglutide solved the dose and efficacy problem. Tirzepatide solved the single-receptor ceiling. Retatrutide is attempting to solve the thermogenic gap. Each solution creates the next problem and the next question.

The endpoint of this iteration isn't obvious from inside it, which is the nature of active science. What is visible from here is that we are somewhere in the middle of a period in which the pharmacology of appetite, metabolism, and energy expenditure is being reconstructed — not by understanding everything at once, but by methodically exploring the tree one branch at a time, in a lizard's venom and a Bronx VA laboratory and a Toronto office and a pharmaceutical campus in Indianapolis, each discovery making the next one legible.