How GLP-1 and GIP Medicines Work as Science and What the Trials Measured

What these drugs are, in two sentences

GLP-1 and GIP medicines are engineered copies of gut hormones that amplify insulin release only when glucose is high, slow the pace at which the stomach empties, and act on appetite centers in the brain. Those three effects are why the drugs lower blood sugar and reduce body weight, and each one shows up in the trials as a separate measurement: HbA1c for glucose control, percent body-weight change for the metabolic effect, and cardiovascular event counts for outcomes that matter to patients. Understanding the biology makes the efficacy claims easier to read, because it tells you which number is measuring which mechanism. This is educational, not medical advice.

The mechanism, one receptor at a time

Your intestine releases two hormones within minutes of a meal: glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP). Both are incretins, meaning they tell the pancreas that food is arriving so insulin release is scaled to the meal. The natural versions are broken down in a couple of minutes by an enzyme called DPP-4. The drugs are built to survive that enzyme, so a single injection keeps the signal running for a week.

The most important word in the mechanism is glucose-dependent. As the mechanistic review summarized in the endocrine literature describes, GLP-1 drives insulin secretion in a glucose-dependent manner, alongside delayed gastric emptying and appetite suppression. In plain terms, the drug only pushes the pancreas hard when blood sugar is elevated. When glucose falls back to normal, the insulinotropic push fades. That built-in brake is why these agents produce strong glucose lowering with low rates of hypoglycemia, a point the SURPASS program documented for the dual agonist tirzepatide.

The receptors sit in more than one place. The review in the primary literature notes that GLP-1 receptors are expressed across the pancreas, brain, gastrointestinal tract, cardiovascular system, kidney, and adipose tissue. Three of those locations drive the effects patients notice. In the pancreas, the drug amplifies insulin and restrains glucagon. In the gut, it slows gastric emptying, so glucose enters the bloodstream more gradually and meals feel filling for longer. In the brain, it acts on appetite-regulating circuits that reduce hunger and food intake.

GIP adds a second axis. Tirzepatide is a dual agonist that engages both the GIP and GLP-1 receptors, and the pairing produces larger effects on glucose and weight than single-receptor agents in head-to-head testing. The newest agents in development, described in the Endocrine Reviews overview, push further by combining GLP-1 with GIP, glucagon, or amylin signaling, which is why you now see triple agonists and amylin combinations in the pipeline. The unifying idea is simple: recruit more of the gut-hormone system at once.

How a biological effect becomes a trial endpoint

Here is the translation step that matters for reading the headlines. Each mechanism maps to a number the trial reports, and the three numbers answer different questions.

HbA1c: the glucose-control number

HbA1c reflects average blood glucose over roughly three months. It is the endpoint that captures the insulinotropic and glucagon-restraining effects. In the SURPASS phase 3 program for tirzepatide in type 2 diabetes, HbA1c fell by roughly 2 percentage points versus placebo across the trials, with the exact figure varying by dose and comparator. When you see an HbA1c reduction, you are seeing the pancreatic mechanism measured.

Percent body-weight change: the appetite and emptying number

Weight loss reported as a percent of starting body weight is largely the readout of reduced appetite and slowed gastric emptying. This is the number that has driven the obesity trials. In the STEP program for semaglutide, sustained weight reduction was reported at 104 weeks, and the SURMOUNT program for tirzepatide reported larger mean reductions still. The direction is consistent and the effect is real, but the exact figure depends heavily on dose, duration, and the population enrolled, which is why two trials of the same drug can report different percentages.

Cardiovascular events: the outcome that matters most

The endpoint hardest to fake and most meaningful to patients is a count of actual events. The SELECT trial, published in the New England Journal of Medicine in 2023, enrolled more than 17,000 adults with overweight or obesity and established cardiovascular disease but without diabetes. Its primary endpoint was a composite of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke, and semaglutide 2.4 mg reduced that composite by about 20 percent relative to placebo. A surrogate like HbA1c tells you the drug moves a lab value. An outcome trial like SELECT tells you it changes what happens to people.

Reading the efficacy claims critically

Once you know which number measures which mechanism, a few habits protect you from over-reading a headline.

First, ask what the endpoint actually was. A reduction in a surrogate marker is not the same as a reduction in heart attacks, and a claim built on a lab value should not be described as if an outcome trial had proven it.

Second, look at the comparator. A drug beating placebo answers a different question than a drug beating another active treatment. The SURPASS-2 trial compared tirzepatide directly against semaglutide, and head-to-head trials like that carry more information than two separate placebo comparisons ever could.

Third, notice the population and duration. Percent weight change and HbA1c both depend on who was enrolled and for how long. A number from a two-year obesity trial and a number from a 40-week diabetes trial are not interchangeable, even for the same molecule.

Fourth, separate relative from absolute effects. A 20 percent relative risk reduction sounds larger than the absolute difference behind it, and both figures are true at once. The honest way to read a trial is to hold them together rather than quote whichever one sounds bigger.

None of this is a verdict on any individual product, and none of it is a recommendation. It is a way to line up the biology and the measurement so the claim in front of you can be judged on what it actually showed.