Tirzepatide and semaglutide are the two best-characterized incretin-based peptides in the metabolic literature, and they are frequently compared. The comparison is more subtle than it looks. The two molecules differ in structure, in which receptors they engage, and in the trials that have tested them. This article separates three things that are often blurred together: what the pharmacology shows, what head-to-head randomized trials show, and what can only be inferred from comparing separate trials.
Two Different Molecules
Semaglutide is a GLP-1 receptor agonist. It is an analog of human GLP-1 with roughly 94% sequence homology, modified with amino-acid substitutions and a fatty-diacid side chain that promotes albumin binding and extends its half-life to about a week (Lau et al., J Med Chem 2015;58:7370-7380). It acts at a single receptor, GLP-1R.
Tirzepatide is a 39-amino-acid peptide built on the native GIP sequence, with a C20 fatty-diacid moiety for albumin binding. It is a dual agonist of GIPR and GLP-1R (Coskun et al., Mol Metab 2018;18:3-14). Because it is single-molecule rather than a combination of two drugs, the relative potency at each receptor is fixed by its structure.
Receptor Pharmacology
In vitro characterization reported that tirzepatide binds GIPR with affinity comparable to native GIP, and binds GLP-1R with lower affinity than native GLP-1. It also shows biased agonism at GLP-1R, favoring cAMP generation over beta-arrestin recruitment, a profile that could influence receptor internalization and signaling duration (Willard et al., JCI Insight 2020;5(17):e140532). Whether this bias contributes to clinical differences is a hypothesis, not an established finding.
The role of GIP is the central mechanistic puzzle. Some data suggest that GIPR agonism enhances insulin secretion and may modulate adipose tissue and appetite pathways; other lines of work have even explored GIPR antagonism as a route to weight loss. A review of how GIP might enhance GLP-1 efficacy (Samms, Coghlan and Sloop, Trends Endocrinol Metab 2020;31:410-421) lays out these competing explanations. No published human trial isolates the GIP contribution, so it remains unresolved.
The Head-to-Head Trials
| Trial | Population | Design | Result |
|---|---|---|---|
| SURPASS-2 (Frías et al., N Engl J Med 2021;385:503-515) | 1,879 adults with type 2 diabetes on metformin | 40 weeks; tirzepatide 5, 10 or 15 mg vs semaglutide 1 mg | HbA1c reduction was greater with all tirzepatide doses (about 2.0 to 2.3 percentage points) than with semaglutide (about 1.9); body-weight reduction ranged from about 7.6 to 11.2 kg versus 5.7 kg |
| SURMOUNT-5 (Aronne et al., N Engl J Med 2025;393:26-36) | 751 adults with obesity, without diabetes | 72 weeks; maximum tolerated tirzepatide (10 or 15 mg) vs semaglutide (1.7 or 2.4 mg) | Mean weight change of 20.2% with tirzepatide versus 13.7% with semaglutide (about 22.8 kg versus 15.0 kg); waist circumference decreased by 18.4 cm versus 13.0 cm |
In SURMOUNT-5, gastrointestinal adverse events leading to discontinuation occurred in 2.7% of tirzepatide participants and 5.6% of semaglutide participants, and most adverse events were mild-to-moderate and clustered during dose escalation. These are the two direct randomized comparisons of the molecules; they are the strongest evidence available for how the two differ in a controlled setting.
Why Cross-Trial Comparisons Mislead
Researchers often line up SURMOUNT-1 (tirzepatide, 72 weeks, mean weight reductions of roughly 15% to 21% across doses; Jastreboff et al., N Engl J Med 2022;387:205-216) against STEP 1 (semaglutide 2.4 mg, 68 weeks, roughly 15%; Wilding et al., N Engl J Med 2021;384:989-1002) and conclude they are similar or different. Differences in duration, baseline characteristics, dose-titration rules, background lifestyle intervention, and analysis populations (treatment-regimen estimand versus efficacy estimand) can easily shift the headline number by several percentage points. The head-to-head trials exist precisely because these comparisons are unreliable.
Limits of the Evidence
- SURMOUNT-5 enrolled adults without diabetes; SURPASS-2 enrolled adults with type 2 diabetes. Results do not transfer cleanly between populations.
- Both trials were open-label or partially open-label in design features (injection devices and dosing differ), a standard limitation of drug-versus-drug comparisons.
- Neither trial tests the mechanistic question of whether GIPR agonism is responsible for the difference.
- Long-term outcomes data (cardiovascular, renal) are being generated separately for each molecule and are not captured here.
Regulatory Status
Both semaglutide and tirzepatide are approved by the FDA as prescription medicines for specific indications. Peptides sold for laboratory use are research materials, are not the approved products, and are not a substitute for them.
Sources
- Aronne LJ, Horn DB, le Roux CW, et al. Tirzepatide as compared with semaglutide for the treatment of obesity. N Engl J Med 2025;393(1):26-36. doi:10.1056/NEJMoa2416394
- Frías JP, et al. Tirzepatide versus semaglutide once weekly in patients with type 2 diabetes (SURPASS-2). N Engl J Med 2021;385:503-515.
- Jastreboff AM, et al. Tirzepatide once weekly for the treatment of obesity (SURMOUNT-1). N Engl J Med 2022;387:205-216.
- Wilding JPH, et al. Once-weekly semaglutide in adults with overweight or obesity (STEP 1). N Engl J Med 2021;384:989-1002.
- Coskun T, et al. LY3298176, a novel dual GIP and GLP-1 receptor agonist for the treatment of type 2 diabetes mellitus: From discovery to clinical proof of concept. Mol Metab 2018;18:3-14.
- Willard FS, et al. Tirzepatide is an imbalanced and biased dual agonist at the human GIP and GLP-1 receptors. JCI Insight 2020;5(17):e140532.
- Lau J, et al. Discovery of the once-weekly glucagon-like peptide-1 (GLP-1) analogue semaglutide. J Med Chem 2015;58:7370-7380.
- Samms RJ, Coghlan MP, Sloop KW. How may GIP enhance the therapeutic efficacy of GLP-1? Trends Endocrinol Metab 2020;31:410-421.
For research use only. Not for human or veterinary consumption. This article is educational and describes published literature; it makes no claims about the use of any product.