What Is an Incretin? The Gut Hormones Behind a Drug Revolution

Your intestine sends your pancreas a memo before the sugar from lunch has finished arriving. The memo is a hormone, and the hormones that carry it are called incretins. The two that matter are GLP-1 (glucagon-like peptide 1) and GIP (glucose-dependent insulinotropic polypeptide). Their basic job is in one line of the paper “Structural basis for ligand recognition of incretin receptors”: agonists of these receptors “stimulate the synthesis and secretion of insulin from pancreatic β-cells.” The gut notices food, releases an incretin, and the pancreas makes and releases insulin.

The “incretin effect” is the name for that extra insulin response, the part you get because the food came in through the gut and the gut said something about it. GLP-1 turned out to do more than talk to the pancreas. It also has roles in “carbohydrate metabolism, and appetite,” and both GLP-1 and GIP help the gut talk to the brain through several signaling pathways. That last part is why a diabetes hormone became the basis of the most crowded corner of obesity drug development. The headline promises thirty years from hormone to drug. The facts I am working from do not lay out that timeline step by step, so what follows explains why the delay makes sense rather than dating each step.

What is an incretin actually paid to do?

It helps to think of the body as a series of employees who get paid for specific things. The gut gets “paid” (in the evolutionary sense, I mean, nobody is cutting checks) for reporting that food has arrived. The pancreas gets paid for getting blood sugar under control. An incretin is the message from the first employee to the second. It works a bit like a text that says “heads up, carbs incoming.”

The receptors that pick up the message have a sort of two-part design. According to the same structural paper, the outer end of the receptor (the N-terminal extracellular domain, which is a fancy way of saying the part sticking out of the cell) handles grabbing the hormone and deciding which hormone counts. The part threaded through the membrane handles actually switching the receptor on. The hormones fold into a helix when they bind, and the main force holding them to the outer domain is hydrophobic interaction (the same basic reason oil droplets clump in water). The receptor has a mailbox and a doorbell. Different parts.

The pancreas also has cells that make glucagon, the hormone that raises blood sugar, and incretins talk to those too. A study on rat pancreatic alpha-cells found that GLP-1 and GIP act on them “via a protein kinase A (PKA)-mediated mechanism to stimulate Ca2+-induced exocytosis.” Translated into one flat sentence: the incretins flip a particular internal switch in the glucagon cells, and that makes them release their contents. Meanwhile GIP, per one industry report, enhances the effects of GLP-1 and increases lipolysis, which is the breakdown of stored fat. So the incretin system is not one memo. It is a group chat.

Why a good hormone is not yet a good drug

Here is the first link in the chain where the incentives stop lining up. The body wants a meal signal to be brief, as a general matter, because meals end and you don’t want the “food is here” alarm ringing all afternoon. A drug company wants nearly the opposite. It wants something that lasts, can be dosed rarely, and gets taken for a long time. Both are acting sensibly. The body is optimizing for lunch. The company is optimizing for a product that someone can actually use on a schedule (and, to be fair, for a product that keeps working, which is the point of medicine).

So the work of turning an incretin into a drug is basically the work of taking a hormone designed to be fleeting and persuading it to stick around. I assume there was some meeting where a chemist said “okay, what if we keep the part the receptor likes and change everything the body uses to get rid of it,” and someone else said “sure, but it still has to fold into that helix.” (This meeting is invented, but some version of it presumably happened many times.) The site’s piece on half-life engineering covers how drugs went from hours to weeks; the short version is that it is hard. That is part of why it took decades.

Durability on paper is not the same as durability in a person’s life, though. A 2026 narrative review comparing real-world and trial evidence found real-world discontinuation rates for incretin-based therapies of 20% to 50% within one year. Trials look better. In the TRIUMPH-2 trial of retatrutide, 965 of 1,152 participants (84%) completed the study drug through week 80. Trials have nurses, reminders and schedules. Real life has the rest.

Once the incretin worked, everyone showed up

The next link: once GLP-1 drugs worked, the incentive for every drug developer became “do that, but more.” A 2024 systematic review, “Emerging pharmacotherapies for obesity,” searched trial registries and databases from 2012 to 2024 and found 53 phase 2 and phase 3 trials covering 36 emerging obesity drugs or combinations (4 of those trials were withdrawn or terminated). Almost half of the obesity drugs in phase 2 as of 2024 were incretin analogs. Completed phase 2 incretin trials showed mean weight loss from 7.4% to 24.2%. When a category works that well, the pipeline stops being a pipeline and becomes a queue.

“More” mostly meant more receptors. Tirzepatide is a 39-amino-acid synthetic peptide that agonizes both the GIP and GLP-1 receptors, which earned it the nickname “twincretin” (pharmacology is not a field known for its restraint in naming). It is, in effect, a molecule built to hook up with two receptors at once, and in the SURMOUNT-1 phase 3 trial it produced mean weight reductions of 19.5% and 20.9% at the 10mg and 15mg doses. The site’s tirzepatide article handles the details.

Then came three. Retatrutide hits GLP-1, GIP and glucagon receptors, and so does UBT251, which in a 24-week phase 2 trial of 205 adults produced weight loss of 13.6% to 19.7%, versus 2.0% on placebo. Adding glucagon is odd on its face, because glucagon raises blood sugar. The trick, per an industry report, is that the dual incretin activity offsets glucagon’s blood-sugar effect while keeping its boost to energy expenditure. You hire the guy who sets fires because he also heats the building, and you hire two firefighters to stand next to him. It is sort of the Ocean’s Eleven approach to receptor pharmacology: every specialist covers for someone else’s flaw.

Where the incentive chain starts to get strange

Follow the money one more link and the science gets confusing. Tirzepatide activates the GIP receptor. Amgen’s maridebart cafraglutide, a bispecific antibody, blocks the GIP receptor while activating GLP-1. Both seem to add weight loss on top of GLP-1. A review titled “The Premise of the Paradox” calls this “a central unresolved scientific question in incretin pharmacology.” Pressing the button helps. Unpressing the button also helps. Nobody has settled why, and the site’s GIP receptor article lays out the competing explanations.

The unresolved list is long, on the other hand, and not just at the receptor level. Reviewers flag gastrointestinal side effects, weight regain after stopping, loss of lean mass, and open long-term safety questions. Effects on gut microbiota “remain uncertain.” Long-term effects in adolescents on growth, puberty and bone are uncertain too. Meanwhile the internet has started calling retatrutide “GLP-3,” a label that does not correspond to any recognized drug class. There is no GLP-3. There is just marketing gravity.

And the chain finally reaches Washington. BioSpace reported that the FDA delayed a decision on Lilly’s obesity pill orforglipron and other awardees of a voucher program, describing it this way: the Commissioner’s National Priority Voucher program “awards tickets to companies that pledge to comply with national priorities such as addressing a key unmet need, boosting domestic manufacturing, and aligning drug prices with President Donald Trump’s Most Favored Nation scheme.” Translated flatly: a faster regulatory review is available to companies that promise, among other things, to price drugs in line with a White House pricing plan. So the gut tells the pancreas about lunch, the pancreas tells chemists about a receptor, the chemists tell investors about a pipeline, the pipeline tells regulators about unmet need, and the regulators tell companies about pricing, manufacturing, etc.

Nobody designed that sequence. Each step was a sensible response to the step before it. Nobody intended this one.

The hormone orforglipron mimics evolved to arrive before the sugar does. By the same logic, a well-designed incretin drug now has to arrive before the pricing policy does, which makes lobbying the next frontier in half-life engineering.