Retatrutide vs. Tirzepatide vs. Semaglutide

Incretins are gut-derived hormones that signal to the pancreas and brain in response to food. Two matter here: GLP-1 (glucagon-like peptide-1) and GIP (glucose-dependent insulinotropic polypeptide). A third receptor, the glucagon receptor, belongs to the same class B G-protein-coupled receptor family and sits on the opposite side of the metabolic ledger. The three compounds below are distinguished by how many of those receptors they engage.

The receptor map

Semaglutide: the reference molecule

Semaglutide is a GLP-1 receptor agonist built on the native GLP-1 backbone with two engineering changes that define it. An alpha-aminoisobutyric acid substitution at position 8 blocks cleavage by DPP-4, the enzyme that degrades native GLP-1 within minutes. A C18 fatty diacid is attached through a glutamate and short polyethylene glycol spacer, which lets the molecule bind reversibly to circulating albumin and extends its half-life to roughly a week.

It is the most extensively characterised molecule of the three. The published literature includes large randomised trials in type 2 diabetes and in obesity, and a cardiovascular outcomes trial in people with overweight or obesity and established cardiovascular disease. In the STEP 1 trial, published in the New England Journal of Medicine in 2021, the treatment group showed a mean body-weight change of −14.9% against −2.4% for placebo at 68 weeks.

Tirzepatide: what adding GIP does

Tirzepatide is built on a GIP peptide backbone rather than a GLP-1 one, engineered so that it activates both receptors — though not equally. Published characterisation work describes it as biased toward the GIP receptor, with GLP-1 receptor activity that is weaker than native GLP-1 and biased away from beta-arrestin recruitment. That is a genuinely unusual pharmacological profile, and it is the reason tirzepatide is not simply 'semaglutide plus something'.

The role of GIP is one of the live arguments in this field. GIP receptor agonism and GIP receptor antagonism have both been reported to produce metabolic effects in preclinical models, and the reconciliation — that sustained agonism may functionally desensitise the receptor — remains a working hypothesis rather than a settled account. Anyone presenting the GIP contribution as understood is overstating the literature.

In SURMOUNT-1, published in 2022, mean body-weight change at 72 weeks ranged from −15.0% to −20.9% across treatment groups against −3.1% for placebo.

Retatrutide: the glucagon receptor question

Retatrutide adds glucagon receptor agonism to the dual profile. On its face this is counterintuitive — glucagon raises blood glucose, which is the opposite of what the other two arms do. The rationale in the discovery literature is that glucagon receptor activation also increases hepatic fat oxidation and energy expenditure, and that pairing it with sufficient incretin activity offsets the glycaemic effect while retaining the expenditure effect. The balance between the three activities is therefore the entire engineering problem.

The published human evidence is phase 2. In the trial reported in the New England Journal of Medicine in 2023, mean body-weight change at 48 weeks reached −24.2% in the highest treatment group against −2.1% for placebo. Phase 3 work was underway at the time of writing; phase 2 results have historically not always been reproduced at phase 3 scale, and retatrutide has not completed that step.

Why the numbers above do not stack into a ranking

It is tempting to line up −14.9%, −20.9% and −24.2% and call it a hierarchy. Those figures come from three separate trials that differ in almost every dimension that matters:

• Duration — 68 weeks, 72 weeks and 48 weeks respectively. Weight-change curves in this class have not fully plateaued at any of those points.
• Population — entry criteria, baseline body weight, and the proportion of participants with type 2 diabetes differ between trials, and diabetes status materially changes the response.
• Placebo response — the comparator arms themselves differ, which shifts the placebo-adjusted difference independently of the drug.
• Lifestyle intervention — the background diet and activity programme accompanying each trial is not identical.
• Analysis convention — treatment-regimen versus efficacy estimands produce different headline numbers from the same dataset.

Cross-trial comparison is a well-known trap in this literature, and it is the single most common error in consumer-facing writing about these compounds. The only reliable comparative evidence comes from trials that randomise participants between two of these molecules directly.

Where direct comparison does exist

Exactly one pair here has been compared head-to-head. SURPASS-2 randomised participants with type 2 diabetes to tirzepatide or semaglutide and reported greater reductions in glycated haemoglobin and body weight in the tirzepatide groups. A later trial, SURMOUNT-5, ran the comparison in participants with obesity and without diabetes and again reported greater mean weight reduction with tirzepatide.

No published trial randomises anyone to retatrutide against either of the other two. Its position at the top of informal rankings rests entirely on cross-trial arithmetic.

Reading these compounds as research material

Two of these three molecules exist as approved pharmaceutical products, which creates a specific interpretive problem. The clinical literature describes those finished products, administered under medical supervision, at controlled specifications. That literature characterises the molecule — its receptor profile, its pharmacokinetics, its measured effects in trial populations — and nothing about it transfers to material supplied for laboratory use.

For laboratory work the relevant attributes are the ones on the certificate of analysis: identity confirmation, purity by HPLC, peptide content by nitrogen determination, and the water content of the lyophilized cake. All three are supplied as lyophilized powder, reconstituted with bacteriostatic water, with lyophilized material stored at −20 °C and reconstituted solution held at 2–8 °C and protected from light.

The short version

• The distinction is receptor count: GLP-1; GIP plus GLP-1; GIP plus GLP-1 plus glucagon.
• Semaglutide has the deepest published dataset, including cardiovascular outcomes; retatrutide's is phase 2 only.
• Tirzepatide's GIP contribution is mechanistically unresolved, not merely additive.
• Only tirzepatide and semaglutide have been compared head-to-head. Retatrutide's ranking is inferred, not measured.
• Clinical trial results describe approved or investigational medicines under supervision, and say nothing about laboratory material.