Tirzepatide sits in the middle of a sequence. Semaglutide activates one incretin receptor, tirzepatide activates two, retatrutide activates three. That framing is useful for orientation and is covered in the complete GLP-1 research guide, but it flattens the interesting part. Tirzepatide does not simply do “GLP-1, plus another one”. It engages its two receptors unequally, and on purpose.
This piece is about the pharmacology behind that asymmetry and what the trial programme reported. Everything below describes published clinical research in patient populations. None of it is a claim about a research material or a suggestion of use.
Two receptors, and the structure that reaches them
Tirzepatide is a single engineered peptide that activates the GLP-1 receptor and the GIP receptor together. GLP-1 is the target the earlier compounds in this class already hit. The addition is GIP, glucose-dependent insulinotropic polypeptide, the other major incretin hormone released by the gut after a meal.
The molecule itself is a 39-amino-acid linear peptide built on the GIP sequence, carrying a C20 fatty acid attached to a lysine near the middle of the chain. That fatty acid is doing pharmacokinetic work rather than receptor work: it lets the peptide bind reversibly and non-covalently to human serum albumin, which slows clearance enough to support once-weekly dosing. It is the same albumin-binding trick used elsewhere in the class, and it is why these compounds are weekly rather than daily.
Imbalanced and biased: what those words mean
This is the part that distinguishes tirzepatide from a simple two-for-one description, and it is worth being precise about, because the two words describe different things.
Imbalanced refers to how strongly it binds each receptor. Structural work published in PNAS found that at the GIP receptor, tirzepatide behaves much like the natural hormone, adopting the same helical conformation and reaching an affinity equalling that of GIP itself. At the GLP-1 receptor the picture is different: affinity is roughly fivefold weaker than natural GLP-1. So the “dual” agonist leans considerably towards the GIP side.
Biased refers to what happens after binding. A receptor can route a signal down more than one pathway, and a biased agonist favours one route over another rather than reproducing the natural hormone’s full response. Work published in JCI Insight characterised tirzepatide as exactly this: it mimics native GIP at the GIP receptor, but at the GLP-1 receptor it favours cAMP generation over β-arrestin recruitment. Reduced β-arrestin recruitment is associated with less receptor desensitisation, meaning the receptor is less inclined to stop responding over time.
Whether that pharmacological signature explains the trial results is a genuinely open question, and the honest answer is that nobody has isolated the contribution. The receptor-level story is well characterised. The causal chain from there to a clinical endpoint is not.
What the diabetes trials reported
Tirzepatide’s first large-scale evidence came from the SURPASS programme, in type 2 diabetes. The trial most often cited is SURPASS-2, published in 2021, which randomised 1,879 patients over 40 weeks and compared three tirzepatide doses against semaglutide at 1 mg weekly.
On the primary glycaemic measure, mean HbA1c fell by 2.01, 2.24 and 2.30 percentage points on tirzepatide at 5, 10 and 15 mg, against 1.86 points on semaglutide. Body weight showed a wider separation, with tirzepatide groups averaging 1.9 kg, 3.6 kg and 5.5 kg below the semaglutide group at the three doses.
One detail matters for reading that comparison fairly: the semaglutide comparator was dosed at 1 mg, which was the approved diabetes dose at the time of the trial. Higher semaglutide doses have since been studied. A comparison is only ever a comparison against the arm that was actually run.
What the obesity trials reported
The obesity programme is SURMOUNT. SURMOUNT-1, published in 2022 and registered as NCT04184622, randomised 2,539 adults with obesity, or overweight plus a weight-related condition, to weekly tirzepatide at 5, 10 or 15 mg, or to placebo, over 72 weeks.
The dose relationship is visible but it is not linear. The step from placebo to 5 mg is far larger than the step from 10 mg to 15 mg, which flattens noticeably. That shape is common in this class and is more informative than the headline number: it suggests the response is approaching a ceiling rather than continuing to climb with dose.
The most common adverse events were gastrointestinal, mostly mild to moderate. That is the pattern across the whole incretin class rather than anything specific to tirzepatide.
The head-to-head trial
Cross-trial comparisons are weak evidence, because different trials enrol different people for different durations. SURMOUNT-5 avoided that problem by randomising participants directly between the two compounds. Published in 2025, it enrolled 751 adults with obesity and without type 2 diabetes, and ran both compounds to their maximum tolerated dose over 72 weeks.
Waist circumference moved in the same direction, averaging 18.4 cm against 13.0 cm. Because both arms were randomised inside one trial, these two figures can legitimately be read against each other, which is not true of most of the numbers that circulate in this area.
Reading these numbers carefully
A few cautions that apply to every figure above:
- Averages are not individuals. A group mean of 20.9% describes the centre of a wide distribution. The trials report that spread, and it is large in both directions.
- Trial conditions are not ordinary conditions. Participants were screened, titrated on a schedule, and supported throughout. Those conditions are part of the result.
- The comparator is part of the claim. SURPASS-2 compared against semaglutide at 1 mg. Change the comparator arm and the difference changes with it.
- Approved status is not the same as settled science. Tirzepatide has an approval and a growing post-approval record, which retatrutide does not. That is a real difference in evidence depth, and it is separate from the size of any reported effect.
In summary
Tirzepatide is the compound where the “count the receptors” shorthand starts to break down. It engages two, but unequally, matching natural GIP at one receptor while binding the other about five times more weakly and signalling through it differently. That asymmetry is well documented at the molecular level and is the genuinely distinctive thing about the molecule.
The trial programme is unusually complete for this class, covering both diabetes and obesity, and including a direct randomised comparison rather than relying on cross-trial arithmetic. What remains unresolved is the link between the receptor pharmacology and the clinical outcomes, which no trial to date has separated out.
For how tirzepatide sits against its neighbours in the class, see the three-way comparison of semaglutide, tirzepatide and retatrutide. For the triple agonist that adds a glucagon receptor on top, see retatrutide and the TRIUMPH programme.