Claims about peptides rest on very different kinds of study, from large drug trials to a handful of experiments in rats, and they are often repeated as if they carried the same weight. They do not. Medical researchers rank evidence by how well a study can rule out chance and bias, and randomised controlled trials in people come first. Regulators such as the FDA approve a medicine only after reviewing results from trials in people. This article explains that ranking and how PeptoKing uses it to grade each peptide.1,
From the laboratory to approval
New medicines follow a set path. Before a drug is tested in people, its developer runs laboratory and animal tests to learn how it works and whether it is likely to be safe. Trials in people follow, usually in three phases: phase I in a small number of healthy volunteers or patients, phase II in a relatively small number of patients with the condition, and phase III in a larger number of patients, to compare the medicine with other treatments and find out how well it works.1,
FDA approval means the agency has reviewed the data and decided that the drug's benefits outweigh its known and potential risks for the people it is meant for. The FDA generally expects results from two well-designed clinical trials, so that the findings of the first are not the result of chance or bias, although for a rare disease one convincing trial may be enough. In the EU, a committee of the European Medicines Agency assesses the application and the European Commission takes a legally binding decision, valid in every EU member state and in Iceland, Liechtenstein and Norway.1,
Why randomised trials count most
In a randomised controlled trial, chance decides who receives the treatment and who receives a comparison such as a placebo. The Cochrane Handbook for Systematic Reviews of Interventions names the two problems that studies without randomisation are open to: confounding and selection bias. That is why it starts evidence from non-randomised studies at a lower level of certainty.4
Cochrane uses the GRADE system, which organisations around the world have adopted to rate the certainty of a body of evidence as high, moderate, low or very low. Evidence from randomised trials starts as high and evidence from observational studies, such as cohort and case-control studies, starts as low. Either can move. Confidence goes down for study limitations, inconsistent results, indirect evidence, imprecision and publication bias, and observational evidence can be rated up, for example when the effect is very large.3,
Why animal results are a weak guide
Animal and cell studies are where drug research starts, not where it ends. A 2004 analysis in the BMJ argued that much animal research into potential treatments for humans is wasted because it is poorly conducted, and gave examples of animal studies that were small and poorly designed, with inconsistent use of randomisation and blinding. Even when animal studies are done well, the authors wrote, how far the results apply to humans remains open to question.5
BPC-157 shows the gap. A 2025 systematic review of BPC-157 in sports medicine included 36 studies, 35 of them preclinical. The single clinical study was a retrospective report in which 7 of 12 patients with long-standing knee pain said their pain improved, and the reviewers found no clinical safety data in humans.9
How PeptoKing grades evidence
Each PeptoKing entry carries one evidence grade. It describes the best human evidence for the main use people take the peptide for, not every claim made about it. The grades follow the same order as GRADE: randomised trials in people first, then smaller or uncontrolled human studies, then work in animals and cells.3
| Grade | What it means | Example |
|---|---|---|
| Strong | Several large randomised controlled trials in people. | Semaglutide: in the STEP 1 trial alone, 1961 adults without diabetes were randomly assigned to semaglutide or placebo for 68 weeks. |
| Moderate | At least one good randomised trial, or consistent smaller human trials. | Retatrutide: a randomised, placebo-controlled phase 2 trial in 338 adults with obesity or overweight found substantial weight loss over 48 weeks. |
| Preliminary | Small human studies, pilots or observational data. | Ipamorelin: a phase 2 trial in 117 patients after bowel surgery, which its authors called small, found no significant difference from placebo. |
| Animal data only | Almost everything known comes from animals or cells. | BPC-157: 35 of the 36 studies in a 2025 systematic review were preclinical. |
Status and evidence are separate
PeptoKing shows regulatory status and evidence grade side by side because one does not imply the other. Elamipretide (Forzinity) received accelerated approval in the US for Barth syndrome on the basis of a trial in 12 patients. It did not beat placebo on the trial's main measures, walking distance and fatigue, and the gains in muscle strength that supported approval appeared only in a longer open-label period. Under accelerated approval, the company continues trials to confirm that the drug works. PeptoKing marks it approved, with a preliminary evidence grade.10
The reverse also happens. Retatrutide already has randomised trial evidence in people, which earns it a moderate grade, but until a regulator approves it, it stays marked as in clinical trials.7
A grade describes what is known about a medicine as it was tested in trials. It says nothing about products sold under the same name outside the regulated supply: the FDA points out that unapproved versions of drugs such as semaglutide and tirzepatide do not go through its review for safety, effectiveness and quality.11