First Head-to-Head Comparison of Semaglutide, Tirzepatide, and Retatrutide Shows Tirzepatide Leading—in Mice

A preclinical study compared all three GLP-1 analogs directly. The results are interesting. They are not yet actionable.

A study published in the International Journal of Obesity has done something the pharmaceutical industry hasn’t: pit semaglutide, tirzepatide, and retatrutide against each other in a direct comparison. The catch—and it’s a significant one—is that the comparison happened in genetically obese mice, not humans. The results are interesting. They are not yet actionable.

What the Study Did

Researchers administered all three GLP-1 analogs to MC4R knockout mice—animals genetically engineered to lack the melanocortin-4 receptor, a key node in the brain’s appetite regulation pathway. MC4R deficiency causes severe obesity in both mice and humans, and mutations in this pathway account for approximately 5–6% of severe early-onset obesity in people. The treatment period was 21 days.

The Results

After 21 days of treatment:

Tirzepatide (dual GIP/GLP-1 agonist): 31.6% body weight reduction.
Retatrutide (triple GIP/GLP-1/glucagon agonist): 24.1% body weight reduction.
Semaglutide (GLP-1 agonist): 19.7% body weight reduction.

All three compounds also improved plasma insulin levels, insulin resistance (HOMA-IR), cholesterol, and markers of liver damage (AST and ALT), and reduced liver hypertrophy.

What This Actually Tells Us

It would be tempting to declare tirzepatide the winner here. The numbers certainly look that way. But three caveats matter.

This is a 21-day mouse study. Human GLP-1 trials run 52–72 weeks. Extrapolating dose-response and efficacy rankings from three weeks in rodents to months in humans is not how pharmacology works.

MC4R-deficient obesity is a specific phenotype. These mice model a particular genetic form of obesity affecting the POMC-MC4R pathway. Results in this population do not automatically apply to the far more common polygenic, diet-induced obesity that accounts for most GLP-1 prescriptions. The Stanford PAM study published this month is a reminder that genetic context matters enormously.

We already have human data for two of these three drugs. Semaglutide and tirzepatide have extensive Phase 3 human trial data. Retatrutide’s TRIUMPH program has reported Phase 3 results showing 28.7% weight loss at 68 weeks in humans. Comparing these drugs’ preclinical performance in a specialized mouse model is scientifically interesting—but it does not supersede the human data that already exists.

What this study does contribute is mechanistic insight: the relative performance of single, dual, and triple agonists in a model where the MC4R pathway is completely knocked out tells us something about which receptor targets matter when that particular regulatory node is missing. For the ~5% of severely obese patients with MC4R pathway mutations, this could eventually inform treatment selection.

For everyone else, the human trials remain the evidence that matters.