Retatrutide Led 3 GLP-1 Drugs on Kidney Fibrosis in Mice
Comparisons between the incretin drugs almost always come from different studies, in different populations, with different endpoints, and then get assembled afterwards by whoever is reading. This paper in iScience, 13 August 2026, did the unfashionable thing of comparing semaglutide, tirzepatide and retatrutide in the same two mouse models of kidney scarring and the same cell line.
The authors report that retatrutide showed the greatest efficacy at the doses tested. The more useful finding is that the three compounds did not appear to get there the same way.
A framing note. This is animal and cell work. Peptra Labs supplies tirzepatide and retatrutide as reference material for laboratory research only, semaglutide is not in our catalogue, and none of this is guidance for human use.
What the retatrutide comparison tested
This is from South-Central Minzu University in Wuhan and the Eighth Affiliated Hospital of Southern Medical University in Foshan.
The authors have narrowed the question down. There is a kidney literature for incretin drugs, but it is almost entirely within diabetic nephropathy, where it is not possible to separate an effect in the kidney from an effect in the diabetes. They went instead to non-diabetic renal fibrosis, where glucose is not a confounder.
The three systems were HK-2 cells, a human proximal tubule line, for the in vitro component. A murine unilateral ureteral obstruction (UUO) model, where one ureter is ligated and the blocked organ scars in a predictable time frame, making it the standard fast model for fibrosis. And an aging mouse model, where scarring accrues slowly and for different reasons.
Using two models rather than one is important here. If a compound works in the obstruction model but not in the aging model, you are learning about mechanism, not potency.
Three drugs, three pathway signatures
In the UUO model, the authors report a different pattern for each drug.
| Compound | Pathways associated with the effect, UUO model |
|---|---|
| Semaglutide | PI3K-AKT inhibition |
| Tirzepatide | PI3K-AKT inhibition and PPAR pathway activation |
| Retatrutide | PI3K-AKT inhibition and NF-kB inhibition, concurrently |
Read that column down and the receptor pharmacology of each molecule is visible in it, worth pausing over.
Semaglutide is a single GLP-1 receptor agonist and shows the single signature. Tirzepatide adds GIP receptor agonism and PPAR activation appears alongside the common pathway. Retatrutide adds glucagon receptor agonism on top of both and what appears is NF-kB inhibition, the canonical inflammatory pathway.
The authors do not claim that the extra receptors caused the extra pathways and neither should readers. Three compounds is not sufficient to attribute a pathway to a receptor. But the ordering is at least consistent and it is the kind of observation that tells the next group what experiment to do.
The aging model flattened the differences
In the aging mouse, all three drugs were associated with mitigation of G2/M arrest, with no comparable separation.
G2/M arrest is worth a sentence. If a cell is stuck in that checkpoint after injury, it does not die and it does not recover either. It is in a persistent state and secretes signalling molecules that drive fibrosis in the tissue around it. In kidney work it is one of the better supported cellular explanations for why an acute injury becomes permanent scarring.
That all three drugs converged there, while diverging in the obstruction model, suggests the common GLP-1 receptor is doing the work in the slow model and the extra receptors are more important in the fast one. It is a hypothesis and the authors are explicit that hypothesis generation is what the paper is for.
What “greatest efficacy” does and does not mean
Retatrutide came out ahead, and the qualifier the authors place on that is not stylistic:Â at the doses tested, in the models studied.
A comparative animal study establishes a ranking at the specific amounts chosen. If one compound was administered higher on its own response curve than another, the ranking reflects that choice as much as the molecules. Nothing in an abstract can tell a reader whether the three were comparable on exposure, on receptor occupancy, or simply on a round number of milligrams per kilogram, and those three choices mean three different things by the word “greatest”.
This is the same caution that applies to every head-to-head and it does not make the result less interesting. It makes it a starting point rather than a destination.
Where the human evidence actually sits
The clinical side of this question moved recently and in a way that gives the mouse work some context.
A prespecified pooled analysis in The Lancet Diabetes and Endocrinology, 7 August 2026, looked at kidney outcomes for semaglutide across the SELECT, FLOW and SOUL trials. This is the level of evidence the class has for the kidney in people, and it exists for semaglutide alone.
There is none comparable for retatrutide because there is no completed outcome trial. That is the honest frame for this paper. A compound that wins a mouse comparison and has no kidney outcome data in people has generated a hypothesis, and the field has been reminded repeatedly what happens to hypotheses of this kind. We reported a matched cohort that found liver outcomes level between tirzepatide and semaglutide over 17 months, in an organ where mechanistic thinking had called for a difference. Mechanism separated. Outcomes did not.
The same dynamic is in the biomarker literature. We reported a lipid biomarker analysis on this compound where the markers moved a lot, and markers are not events.
What this is worth to a laboratory
For groups who work on fibrosis models rather than metabolic ones, the practical content of this paper is the protocol.
It establishes that HK-2 cells plus UUO plus an aging cohort will discriminate between three closely related agonists, and it names which readouts did the discriminating. That is a usable experimental design, and designs are more portable across labs than conclusions are.
Preparation identity carries the usual weight and slightly more than usual in a comparative study. If three compounds are being ranked against each other, a difference in whatever is actually in one of the three vials becomes a difference in the ranking and nothing downstream of that will reveal it. Our research reference hub has the identity and purity fields worth confirming first, the triple-agonist research guide covers what the published record supports for this compound, and the European research buyer guide covers procurement for European laboratories.
Everything is supplied on a research use only basis.
References
- Ding M, Li X, Wei Y, et al. Comparative effects of semaglutide tirzepatide and retatrutide on renal fibrosis in UUO and aged mice. iScience, 13 August 2026. doi 10.1016/j.isci.2026.117174
- Mann JFE, Badve SV, Baeres FMM, et al. Effect of semaglutide on kidney outcomes in the SELECT, FLOW, and SOUL trials: a prespecified pooled analysis. The Lancet Diabetes and Endocrinology, 7 August 2026
- Tian X, Ma X, Song E, Li X. From FLOW to translational positioning in chronic kidney disease: mechanistic complementarity of SGLT2 inhibitors and GLP-1 receptor agonists. Translational Research, 21 July 2026
The products referenced on this site are supplied for laboratory research use only. They are not medicines and are not intended for human or veterinary use. This article summarises published research for informational purposes and is not medical advice. Statements about third-party studies belong to their authors.
