Peptide basics

How to check any peptide claim yourself in three minutes

This is a research method, not medical guidance. It tells you what the public record holds, which is a different question from whether anything is safe or appropriate for a person.

Minute one: has anyone registered a human trial?

Go to ClinicalTrials.gov and search the compound name in the intervention or treatment field rather than the general search box. The URL pattern is clinicaltrials.gov/search?intr=NAME, so you can go there directly.¹

Then do the part almost nobody does: open the results. The search is fuzzy by design, and a hit is not a trial of your compound until you have seen the compound in the study's intervention section. Two checks make the difference between a real count and a made-up one.¹²

1. Confirm the compound appears in the interventions list, not merely in the title, the condition, or the eligibility text. A study of something else that mentions your compound in passing is not a study of your compound.
2. Expand the other-names field on each intervention. Development compounds are routinely registered under a sponsor code, and the common name sits there.¹²

Both traps are live. A search for MGF returns over a hundred studies, almost all of them concerning African swine fever virus genes or unrelated drugs; none names mechano growth factor as an intervention. In the other direction, NCT02637284 registers its intervention as Bepecin, with BPC-157 and PCO-02 in the other-names field — a real trial of the compound that a name-only search will miss.¹⁹

If the literature is not English-language

ClinicalTrials.gov is one registry. The WHO's ICTRP search portal queries the network's primary registries together and is the right second stop for compounds whose research base is outside the United States. It is free and needs no account.⁷

Minute two: is there an approved product?

Drugs@FDA is the searchable database of FDA-approved drug products. Search the active ingredient name. If a product exists you will get an application number — an NDA, ANDA or BLA — a sponsor, a brand name, and a marketing status.³

Read four fields and stop. The application number, because it is the stable identifier. The approval date, because a pre-2000 approval will have no registry trail at all. The marketing status, because Discontinued is common and does not mean withdrawn for safety. And the active ingredient, because brand names and ingredient names diverge.³

If you would rather query it programmatically, openFDA exposes the same data as a public JSON API with no key required for light use. A request to the drugsfda endpoint filtered on an application number returns the products, the active ingredients and the submission history in one response.⁴

Minute three: what kind of literature exists?

A bare PubMed search returns a number that flatters almost any compound. Two additions fix it. Put the name in quotation marks followed by [tiab], which restricts matching to the title and abstract. Add a date range in [dp]. The result is a count of records that are actually about the term, over a period you chose.⁵⁶

The difference is not cosmetic. An unfiltered search for MGF returns 2,001 records; the same term restricted to title and abstract over the last ten years returns a few hundred, and most of those still concern unrelated uses of the letters. Comparing equal-length date windows — 2017 to 2019 against 2023 to 2025, say — also shows direction, which a single total cannot.⁵

Then read study types rather than counting. On any result page, the publication type is listed on the record, and the abstract's first methods sentence names the subjects. Rodent, cell line and review records are all common in peptide literatures and none of them is a human trial.⁵

The optional fourth check

For anything sold as a compounded preparation in the United States, the FDA publishes its assessments of bulk drug substances nominated for compounding, with the date of any category placement and the agency's stated reason. It is a single page, it is searchable by substance name, and for peptides it is frequently more specific than anything else available.⁸

Read the reasons literally. For several peptides the agency's stated position is that it has identified no human exposure data and therefore lacks the information needed to know whether the substance would cause harm. That is a description of a gap, and it should not be reported as a finding of danger — nor as a clearance.⁸

What three minutes can and cannot buy you

At the end you can say, accurately: as of today, this registry holds this many studies naming this compound as an intervention; this database does or does not hold an approved application; and the literature is of this size and mostly of this kind. That is a complete description of the public record.¹³⁵

It is not a description of the compound. It cannot tell you what a molecule does in a body, whether anything is safe, or whether any product contains what its label says. What it does is make one specific kind of claim impossible to sustain: the claim that evidence exists, made by someone who has not looked.