Research
Animal studies
What preclinical animal work can and cannot tell us about how a compound will behave in humans.
- Last updated:
- 2026-07-22
Overview
Animal studies (preclinical research in rodents, dogs, primates, or other species) generate the first evidence that a peptide reaches its target, produces a measurable effect, and can be tolerated at defined doses. They are the gate through which every therapeutic candidate passes before human trials.
They are also fundamentally limited. Animal models differ from humans in metabolism, receptor distribution, immune response, lifespan, and behaviour. A compound that behaves well in a mouse may behave differently in a human, and vice versa.
Why it matters
Level D on Healthy Mango's evidence hierarchy means the available evidence is preclinical. That is not the same as "no evidence" — it is a specific kind of evidence with specific limits.
Key concepts
Predictive value varies
Different endpoints translate from animals to humans with different reliability. Toxicology signals are more portable than efficacy signals; behavioural findings translate less reliably than biochemical ones.
Species matters
Rodent, canine, and primate studies produce different reliability of extrapolation. Regulators typically require multiple species before human trials.
Dose scaling
A dose in mg/kg in a mouse is not the same physiological exposure as the same mg/kg in a human. Allometric scaling and pharmacokinetics adjust the translation.
Common misconceptions
- "It worked in mice" is not evidence that it will work in humans. It is evidence that a human trial is worth designing.
- "It was safe in animals" is not evidence that it will be safe in humans. Some human-specific toxicities cannot be detected in animals.
- "Animal-only evidence is worthless." It is not; it is the necessary starting point. But it is starting-point evidence, not endpoint evidence.
Practical interpretation
Where a compound page describes results from animal studies only, treat them as pharmacological orientation. Look for human trials before drawing conclusions about human effect.
Limitations
Not every animal study is well-designed. The number of animals, the endpoints chosen, the blinding, and the pre-registration all matter. Compound pages cite the specific studies for readers who want to look closer.