Independent · Evidence-led · We don't sell peptides
EU / NordicsUpdated weeklyEN

Explore this article's sources with AI

ChatGPTClaudePerplexityGeminiGrokGoogle AI

Follow PeptideMethods on Google

First published

Semaglutide and bone loss in rats

A female-rat study linked semaglutide and PYY3-36 to worse trabecular bone measures, but it cannot establish a human fracture risk.

Why we wrote this. Animal bone findings need a visible species boundary before they are interpreted as evidence about patients.

In this article (5 sections)
  1. How the rat experiment was set up
  2. What changed in weight and bone
  3. What the molecular results can and cannot say
  4. Why this does not establish a human risk
  5. What we do not yet know

A 2026 experiment found that semaglutide and peptide YY3-36, or PYY3-36, were associated with worse measures of the spongy bone inside the femur in female Wistar rats with experimentally induced metabolic osteoporosis. Semaglutide produced a modest change, while the paper reported more pronounced deterioration with PYY3-36 alone and with the combination. This was an animal study, not a human clinical trial. It cannot show that semaglutide causes osteoporosis or fractures in people[1].

How the rat experiment was set up

The researchers randomized female Wistar rats to five groups: a healthy control; metabolic osteoporosis with saline; metabolic osteoporosis with semaglutide; metabolic osteoporosis with PYY3-36; and metabolic osteoporosis with both compounds. Rats in the four metabolic-osteoporosis groups received a Western diet plus L-NAME, a nitric oxide synthase inhibitor used here as part of the disease model. Treatment allocation occurred in week 9, and the investigators collected femora after 16 weeks[1]. The indexed abstract does not report the number of animals or either treatment dose, so this summary does not supply them.

The team examined the femora using micro-computed tomography, a three-dimensional X-ray method for measuring bone structure, alongside histology. They used RNA sequencing and quantitative PCR to assess gene expression. Blood serum was assessed for bone-turnover markers and steroid profiles. These measurements can describe bone architecture and cell counts. They can also identify molecular signals in this model. They are not the same as measuring fractures or patient reports about pain and mobility[2]. Human evidence and approved-use questions belong in the separate semaglutide evidence overview.

What changed in weight and bone

Semaglutide alone and semaglutide combined with PYY3-36 significantly reduced body weight. The abstract does not say that PYY3-36 alone significantly reduced weight. It also does not provide numerical weight changes or confidence intervals, leaving too little detail to compare the size of the two reported weight effects[1]. That missing information matters because a statistically significant change does not by itself tell readers how large or biologically important it was.

For bone, semaglutide produced what the authors called a modest reduction in trabecular bone volume fraction. Trabecular bone is the lattice-like bone inside structures such as the end of the femur, and bone volume fraction describes how much of the measured region is bone. PYY3-36 alone and the combination produced more pronounced trabecular deterioration. The combination also increased trabecular separation, meaning the spaces between the small bone struts were larger. Cortical bone parameters, which describe the denser outer shell, were unaffected in the reported analysis[2].

The abstract reports increased numbers of osteoclasts, cells that resorb or break down bone, in the semaglutide and PYY3-36 groups. It does not provide the counts or effect estimates. Nor does it report an animal fracture endpoint. The result is therefore a structural and cellular signal within this rat model, not evidence that either compound increased human fracture risk[1]. Readers looking for the medicine's current human regulatory context can use our US semaglutide regulation guide, which is separate from this animal experiment.

What the molecular results can and cannot say

Gene-expression changes in femora were mild in semaglutide-treated rats. PYY3-36 treatment significantly increased expression of genes associated with angiogenesis, the formation of blood vessels, and the extracellular matrix, the material surrounding cells. Those associations may help researchers choose mechanisms to test next, but they do not prove that the gene changes caused the trabecular findings. The abstract does not present an intervention that blocks those pathways and reverses the bone changes[2].

The paper's conclusion is strongest when kept inside its tested setting: female rats, a Western-diet and L-NAME model, pre-existing metabolic bone disease, and a 16-week endpoint. The experiment was designed to probe skeletal effects under those conditions. It was not designed to estimate the balance of benefits and risks for a person prescribed semaglutide, and it did not compare clinical osteoporosis treatments[1]. No treatment decision should be based on this experiment alone.

Why this does not establish a human risk

Animal models can expose signals that deserve follow-up, but species and disease stage affect translation. Model induction, exposure, and skeletal biology do too. Here the metabolic-osteoporosis state was created experimentally with diet and L-NAME. That is not interchangeable with osteoporosis arising across years in people with different ages or hormone status and varied patterns of medicine use, nutrition, and activity. The use of female rats also does not make the result a clinical study in women[2].

The study also cannot separate a direct drug effect on bone from every consequence of altered weight, food intake, metabolism, or the induced disease model on the information exposed in the abstract. Semaglutide and the combination reduced body weight, but the indexed record does not report mediation analyses showing how much of the bone finding was explained by weight change. Calling the result a proven direct skeletal toxicity would therefore go beyond the accessible evidence[1].

What we do not yet know

The accessible record leaves important details unresolved: group sizes, doses, food intake, full numerical effects, uncertainty around those effects, allocation concealment, blinding, adverse events, and whether the findings persisted after treatment. It also does not establish whether the combination effect was additive, interactive, or driven mainly by PYY3-36. The full article was not available through PubMed Central, Europe PMC, or the publisher page during this review, so claims here stay within the indexed abstract and metadata[2].

The useful takeaway is narrow. This rat study raises a reason to investigate skeletal outcomes when gut-hormone approaches are tested in the setting of existing bone disease, especially for combinations involving PYY3-36. It does not overturn human evidence, quantify a clinical fracture risk, or support stopping or changing a prescription. Anyone concerned about osteoporosis or a medicine's effect on bone should discuss their individual risk with a qualified clinician[1].

Frequently asked

Was this a semaglutide clinical trial in people?

No. It was an experiment in female Wistar rats with metabolic osteoporosis induced using a Western diet and L-NAME. It cannot establish osteoporosis or fracture risk in people.

What part of the rats' bones changed?

The paper reported deterioration in trabecular bone, the lattice-like inner structure, including a modest reduction in bone volume fraction with semaglutide and more pronounced deterioration with PYY3-36 and the combination. Reported cortical bone parameters were unaffected.

Should someone stop semaglutide because of this study?

No treatment decision should be made from this animal study alone. It does not quantify human fracture risk or test clinical care. Concerns about bone health and a prescription medicine should be discussed with a qualified clinician.

Sources

  1. [1]Kloock S et al. Weight loss at a cost: Semaglutide and PYY3-36 aggravate pre-existing metabolic osteoporosis in female Wistar rats. Bone. 2026. PMID 43025770Tier 1 · primary↩
  2. [2]NCBI PubMed EFetch record for PMID 43025770, including indexed abstract and metadataTier 1 · primary↩

No revisions yet. First published .

About the editorial team

PeptideMethods is written and edited by the PeptideMethods Editorial Team and published by Digital Compass Group Ltd. The team is not made up of medical professionals; every health, regulatory or dosage claim on the site is tied to a primary source and is not a substitute for advice from a qualified clinician.

See our editorial policy and methodology for how we research, source and verify.

Read the pillars