METABOLIC & WEIGHT RESEARCH / SIDE BY SIDE
Three Peptides, Three Positions on the Energy-Balance Equation
Where each compound acts, what was actually measured, how mature the evidence is, and where the record stops.
The short version
This page sets MOTS-c, retatrutide and semaglutide against each other on the dimensions that separate them: which side of the energy balance each one acts on, what kind of evidence exists, and how far along the regulatory road each one is.
The pattern is straightforward once it is laid out. Semaglutide acts on intake and has by far the deepest evidence, including hard outcome trials in tens of thousands of people. Retatrutide acts on intake and expenditure and shows the largest weight numbers, but every one of those numbers comes from Phase 1 or Phase 2, and it is approved nowhere. MOTS-c acts inside the cell on how fuel is handled, has the most novel biology of the three, and has no completed human efficacy trial at all.
Bigger mechanism, thinner record; thinner mechanism, bigger record. That inverse relationship is the single most useful thing to carry away from this comparison. None of it is medical advice, and no dose here is a recommendation for any person.
The comparison table
| Dimension | MOTS-c | Retatrutide | Semaglutide |
|---|---|---|---|
| Class | Mitochondrial-derived signalling peptide | GIP/GLP-1/glucagon triple receptor agonist | GLP-1 receptor agonist (incretin mimetic) |
| Side of energy balance | Storage and cellular fuel handling | Intake and expenditure | Intake |
| Primary target | AMPK, via folate cycle; CK2 directly [1] | GLP-1R, GIPR, GCGR [7] | GLP-1R, centrally and in the pancreas |
| Best weight evidence | None in humans | -24.2% at 48 weeks, 12 mg, Phase 2 [9] | -14.9% at 68 weeks, 2.4 mg [16] |
| Hard clinical outcomes | None | None yet; trials ongoing | MACE HR 0.80 [15]; kidney composite HR 0.76 [14] |
| Human evidence maturity | Observational biomarker association only [2] | Phase 1 and 2 complete, Phase 3 ongoing | Multiple completed outcome trials |
| Administration studied | Rodent studies only | Once-weekly subcutaneous, half-life ~6 days [11] | Once-weekly subcutaneous; once-daily oral |
| Regulatory status | Not approved; research chemical | Investigational; not approved anywhere | FDA-approved, multiple indications |
| Anti-doping | Prohibited in elite sport | Not specifically prohibited; verify current list | Not specifically prohibited; verify current list |
| Principal caution | Human effects unknown | Unapproved; GI events and heart rate [9] | GI intolerance; class warnings [17] |
Where each one acts
Semaglutide — intake only. GLP-1 receptor agonism in the hypothalamus and brainstem reduces food intake and shifts food preference, with slowed gastric emptying reinforcing it. The published mechanism is explicit that energy expenditure is not lowered, so the effect is a clean reduction on one side of the ledger.
Retatrutide — intake and expenditure. The GLP-1 and GIP arms do the intake work; the glucagon arm adds energy expenditure and lipid mobilisation. The structural work shows the arms are deliberately unbalanced — roughly 8.9-fold more potent than native GIP at GIPR, but only 0.3-fold and 0.4-fold as potent as the endogenous hormones at GCGR and GLP-1R [7] — a restrained glucagon signal riding alongside strong incretin signalling. The metabolomic post-hoc analysis is the closest thing to direct evidence for the expenditure route: fatty-acid-oxidation changes mediated 23.2% of the weight-reduction response in participants without type 2 diabetes [12].
MOTS-c — beneath both. No appetite mechanism, no whole-body expenditure mechanism. AMPK activation via folate-cycle inhibition, muscle glucose uptake, CK2 as a direct binding target, and nuclear translocation under metabolic stress [1][5]. This is a story about the efficiency with which a cell handles fuel, which is a different question from how many calories cross the mouth or leave as heat.
Evidence maturity, honestly ranked
Semaglutide. Completed outcome trials with hard endpoints: 17,604 adults for cardiovascular events [15], 3,533 for kidney events [14], 1,961 for weight [16], plus a head-to-head against tirzepatide in 751 adults [13] and a dedicated safety synthesis [17]. Approved labelling exists, which means an independent regulator has reviewed the underlying data.
Retatrutide. Phase 1b in 72 adults [11], Phase 2 in 338 adults with obesity [9], Phase 2 in 281 adults with type 2 diabetes [10], a Phase 2a imaging study in 98 participants with metabolic liver disease [8], and post-hoc metabolomics across 495 participants from two of those trials [12]. Large effects, real trials, mid-sized populations, and no completed Phase 3 or outcome trial. No regulator has reviewed it for approval.
MOTS-c. Cell-free assays, mouse work across several ages and models [1][4], a mechanistic translocation study in human and mouse cells [5], a 2023 review consolidating the field [3], and one prospective human cohort of 94 haemodialysis patients in which circulating levels were associated with a mortality and cardiovascular composite [2]. That last study measures the body's own MOTS-c as a marker; it does not test giving MOTS-c to anyone.
Three different questions are being answered at these three levels of evidence — does this improve outcomes, does this move a surrogate endpoint, and does this mechanism exist — and reading a level-three answer as though it were a level-one answer is the most common error in this subject.
How to read the numbers on this desk
Percentages compare badly across trials. STEP 1 ran 68 weeks in adults with overweight or obesity without diabetes [16]; the retatrutide Phase 2 obesity trial ran 48 weeks in a population selected on different criteria [9]; SURMOUNT-5 ran 72 weeks and is the only head-to-head in this set [13]. Duration, population, comparator and escalation schedule all move the headline figure, so a difference between two trials is not the same kind of fact as a difference within one.
Endpoint class matters more than effect size. A hazard ratio for major adverse cardiovascular events in 17,604 people [15] is a different order of evidence from liver fat on a scan [8], which is in turn different from a biomarker association in 94 people [2] — even though all three can be written as a percentage in a summary paragraph.
Where a dose appears anywhere on this site, it is a description of what a study administered, given so that a figure can be attached to the conditions that produced it. It is never a protocol, never a recommendation, and never an implication that the same conditions could be reproduced outside a trial. Retatrutide and MOTS-c are not approved products; semaglutide is a prescription medicine whose use is a matter between a patient and a licensed prescriber.