Quick answer
What do rodent studies of the cagrilintide–semaglutide combination (CagriSema) show about how to design a combination experiment? The best-documented work separates a combination’s effect on energy intake from its effect on energy expenditure by adding pair-fed and weight-matched control groups. In one rat study the combination produced 12% body-weight reduction with a 39% fall in food intake, pair-feeding produced less, and weight matching needed a 51% intake cut, which the authors read as roughly one-third of the effect arising from expenditure. This article explains factorial and add-on designs, synergy analysis and reporting. It is not a use guide. For laboratory research use only.
CagriSema is the shorthand for a combination of two separate molecules: cagrilintide, an amylin and calcitonin receptor agonist, and semaglutide, a GLP-1 receptor agonist. In the laboratory the combination is not one reagent but two, supplied and handled separately, and every design decision about it is a decision about how two sets of variables interact. This article approaches the combination as an experimental design problem: what the rodent literature asked, what controls it used, how synergy is defined and what a clean report contains. Both peptides are sold strictly for laboratory research use only; not for human or animal consumption.
Why two receptor systems are paired
The scientific logic is complementarity. Amylin acts largely through brainstem circuits that end meals, with the area postrema and nucleus of the solitary tract implicated (Lutz, 2012). GLP-1 receptor agonists act through a broader set of central and peripheral sites. Semaglutide itself is an acylated GLP-1 analog designed for albumin binding and long persistence (Lau and colleagues, 2015). Two molecules with different primary sites and a shared strategy of lipidation make a natural pair for questions about whether signals add up. The longer background on each compound is in the cagrilintide research guide and the semaglutide research guide, and the earlier combination overview is in the CagriSema combination research article.
Three design types for a two-compound study
| Design | Arms | What it answers | What it cannot answer |
|---|---|---|---|
| Add-on | Vehicle; compound A; A plus compound B | Does B add to an established A effect? | Whether B alone has an effect; whether A adds to B |
| Full factorial | Vehicle; A; B; A plus B | Additivity vs synergy vs antagonism | Dose-ratio dependence unless several ratios are run |
| Fixed-ratio dose-response | Several combined levels at one molar ratio | Potency shift of the combination relative to the singles | Which ratio is optimal |
The published rodent work uses these structures in different ways. A characterization of a cagrilintide tool analog described "additional add-on effects on top of semaglutide" in diet-induced obese mice and rats (Gabery and colleagues, 2025), an add-on design. A 2026 study of cagrilintide with retatrutide compared the combination with equimolar monotherapies and with matched-dose combinations using semaglutide or tirzepatide (Petersen J and colleagues, 2026), which is a factorial-style arrangement with additional reference combinations. Which one a lab uses depends on the claim it wants to make; an add-on design supports only the claim its arms permit.
What "synergy" requires
Many reports call a combined effect "synergistic" simply because it exceeds one of the single-agent effects. That is not synergy. A defined reference model is needed. Reviews of combination analysis separate effect-based approaches from dose-effect-based approaches (Foucquier and Guedj, 2015); two widely used references are Loewe additivity, which uses equally effective amounts of each agent, and Bliss independence, which models the agents as acting independently. Each predicts an expected combined effect from the single-agent data, and synergy is a combined effect larger than that prediction, with antagonism the reverse (Tallarida, 2001). The consequence is practical: without single-agent dose-response data, the expected additive effect cannot be calculated, so an add-on design cannot support a synergy claim. Where only a fixed combination was tested, the honest wording is "the combination reduced intake by X"; whether this exceeds additivity is not answered.
Separating intake from expenditure: the rat calorimetry design
The most informative recent design for this pair addresses a different question: when body weight falls, how much of the change is explained by reduced intake and how much by energy expenditure? In a rat study, indirect calorimetry was combined with two calorie-restriction controls. A pair-fed group received the same food as the exposed animals. A weight-matched group was restricted until its body weight matched the exposed group. The authors reported 12% body-weight reduction with a 39% reduction in food intake for the combination, less pronounced reduction with pair-feeding, and the need for a 51% food-intake decrease to reach the same weight by restriction, from which they concluded that about one-third of the efficacy arose from an effect on expenditure, a blunting of metabolic adaptation (Jacobsen and colleagues, 2025).
A follow-up reported four phases (Petersen LM and colleagues, 2026): an initial phase of rapid reduction, mainly through lower energy intake with expenditure relatively preserved; a second phase in which reduced body weight was maintained in the combination, weight-matched and pair-fed groups, with respiratory exchange ratios converging; a third phase after discontinuation with hyperphagia and rapid regain; and a fourth phase of return to the pre-exposure rate of weight gain. In that study reduced intake and preserved expenditure respectively explained 86% and 14% of the efficacy. The two papers give different attribution numbers because they ask the question over different windows, a reminder that a decomposition depends on when it is measured.
| Control group | Purpose | Reading |
|---|---|---|
| Vehicle | Baseline trajectory and handling effects | Reference for all comparisons |
| Pair-fed | Same intake as exposed group | Isolates effects not explained by intake |
| Weight-matched | Same body weight by restriction | Shows metabolic adaptation in the absence of the compounds |
| Single-agent arms | Contribution of each component | Needed to judge additivity |
| Discontinuation arm | Rebound and recovery of adaptation | Shows persistence of effect |
Statistics for a multi-arm combination study
A factorial layout is analyzed with a two-way model: compound A (present or absent), compound B (present or absent) and their interaction term. A significant interaction is evidence that the effects are not simply additive on the scale analyzed, but the scale matters. An effect that is additive on percentage change can look interactive on absolute change, so the analysis scale should be stated in advance. Repeated body-weight and intake measurements on the same animals call for a repeated-measures or mixed model with animal as a random effect, not separate t-tests at each day. Sample size should come from a power calculation based on a pilot or on prior variance, and reporting guidelines for animal research expect the calculation, the unit of analysis and the handling of exclusions to be stated. Where several comparisons are run, the correction method belongs in the methods, not the discussion. Report effect sizes with confidence intervals alongside p-values so that readers can judge magnitude and not only significance.
Handling two peptides in one study
Because the combination is two molecules, handling discipline doubles. Each stock needs its own identity record, its own concentration calculation and its own storage history. The vial masses, solvents and final molar concentrations should be computed separately with c = m ÷ M ÷ V and verified in the reconstitution calculator. If the two peptides are to be co-administered, the report should say whether they were mixed in one solution or given separately, because mixing can change adsorption and aggregation behavior; this is an experimental variable, not a given. Cagrilintide is amphiphilic and amylin-derived, which makes it more sensitive to shaking, freeze–thaw and surface adsorption than many peptides; the practical reasons are laid out in how long reconstituted peptides last and storing research peptides. The reagents are listed as the cagrilintide vial and the semaglutide research peptide, each with product-page storage and purity statements.
A reporting checklist for combination studies
- Molar amounts of each component in every arm, and the molar ratio of the combination.
- Whether combination arms were matched to single-agent arms by molar or mass basis.
- Single-agent arms, or an explicit statement that additivity was not tested.
- Pair-fed and weight-matched controls where intake and expenditure are discussed.
- Time window for every attribution figure, since percentages shift with the window.
- Sex, strain, diet and housing, as a recognized animal-research reporting guideline asks.
- Lot numbers and analytical records for each peptide, from the COA library where published.
Several of the abstracts cited here describe male rats, so any sex difference is outside what they report.
Limits of the evidence
The rat energy-balance studies cited were produced by groups associated with the developer of the combination, and several authors declare that employment or shareholding. That does not invalidate the work, but independent replication in other laboratories would strengthen conclusions about attribution percentages. The studies are also in male diet-induced obese rats, so they do not establish what happens in other species, strains, ages or sexes. None of this applies outside laboratory research, and the article makes no recommendation about amounts or schedules. Related reading: cagrilintide vs tirzepatide, GLP-1 receptor agonists in research and the research catalog.
For laboratory research use only.
Frequently asked questions
Is CagriSema one compound?
No. It is shorthand for a combination of two separate molecules, cagrilintide (an amylin and calcitonin receptor agonist) and semaglutide (a GLP-1 receptor agonist). In the laboratory they are separate reagents.
What did the 2025 rat study report for the combination?
The authors reported 12% body-weight reduction with a 39% reduction in food intake, less reduction with pair-feeding, and a 51% intake decrease needed to match the weight by restriction, implying about one-third of efficacy from expenditure effects.
Why use pair-fed and weight-matched controls?
A pair-fed group shows what the same intake alone does. A weight-matched group shows what restriction to the same body weight does, including metabolic adaptation. Together they help attribute effects to intake or expenditure.
Why do different studies give different intake and expenditure percentages?
Attribution depends on the window measured. A 2026 follow-up reported 86% from reduced intake and 14% from preserved expenditure over its exposure period, a different window from the earlier one-third estimate.
What is the difference between additive and synergistic?
Additivity is a calculated expectation from single-agent data using a reference model such as Loewe or Bliss. Synergy is a combined effect larger than that expectation.
Can an add-on design show synergy?
No. Without single-agent dose-response data for both components, the expected additive effect cannot be calculated.
What happened after exposure ended in the rat study?
The follow-up reported hyperphagia and rapid weight regain after discontinuation, then a return to the pre-exposure rate of weight gain.
Should the two peptides be mixed?
That is an experimental variable. The report should state whether they were mixed or given separately, since mixing may change adsorption and aggregation behavior.
Are the cited studies independent of the developer?
Several authors of the cited energy-balance papers are employed by the developer and declare it. Independent replication would strengthen the conclusions.
Does this article recommend amounts or schedules?
No. It describes design and reporting for laboratory research only.
Related reading: Cagrilintide Animal Research Protocols: Design & Math, Cagrilintide Safety Profile: Preclinical Tolerability Data, Cagrilintide vs Tirzepatide: Assay Design and Data.
References
- Jacobsen JM, Halling JF, Blom I, et al. CagriSema drives weight loss in rats by reducing energy intake and preserving energy expenditure. Nat Metab. 2025;7(7):1322-1329. doi:10.1038/s42255-025-01324-8
- Petersen LM, Knop FK, le Roux CW, Raun K, Reitman ML, Kuhre RE. Continued CagriSema treatment preserves weight loss by reducing energy intake and maintaining energy expenditure in rats. Mol Metab. 2026;115:102464. doi:10.1016/j.molmet.2026.102464
- Gabery S, Glendorf T, Ballarin-Gonzalez B, et al. Characterization of 0839, a tool compound for pre-clinical mode-of-action studies of amylin analogues such as cagrilintide. Life Sci. 2025;378:123845. doi:10.1016/j.lfs.2025.123845
- Petersen J, Merrild C, Holm SK, et al. Cagrilintide and retatrutide combination therapy enhances weight loss and metabolic outcomes in obese male rats. Nat Metab. 2026;8(10):2069-2083. doi:10.1038/s42255-026-01603-y
- Lau J, Bloch P, Schäffer L, et al. Discovery of the once-weekly glucagon-like peptide-1 (GLP-1) analogue semaglutide. J Med Chem. 2015;58(18):7370-7380. doi:10.1021/acs.jmedchem.5b00726
- Tallarida RJ. Drug synergism: its detection and applications. J Pharmacol Exp Ther. 2001;298(3):865-872. doi:10.1016/S0022-3565(24)29450-7
- Foucquier J, Guedj M. Analysis of drug combinations: current methodological landscape. Pharmacol Res Perspect. 2015;3(3):e00149. doi:10.1002/prp2.149
- Lutz TA. Control of energy homeostasis by amylin. Cell Mol Life Sci. 2012;69(12):1947-1965. doi:10.1007/s00018-011-0905-1
