Verified July 2026 · The methodology behind every grade
How we grade peptide evidence.
Every peptide gets a letter grade, A to F, set by a fixed 6-rule decision tree, ranked by the strength of its human evidence: from FDA-approved at A to failed-in-humans at F. The grade is mechanical, and it is never influenced by whether we can monetize the peptide.
What is the exact rule for each grade?
We run each peptide top-to-bottom through this decision tree and stop at the first rule that fires. That is the grade. No discretion, no “feel,” no exceptions for compounds we could earn a commission on.
- 1FDA-approved for ≥1 indication?→ Grade A
- 2Else, ≥2 published Western peer-reviewed human RCTs?→ Grade A
- 3Else, ≥1 real human clinical trial OR prior FDA approval OR foreign approval backed by Western-recognizable trials?→ Grade B
- 4Else, human pilot/observational data OR foreign-only (e.g. Russian) approval with weak RCTs?→ Grade C
- 5Else, only animal/preclinical, no published human efficacy RCT?→ Grade D
- 6Else, a human trial exists and was negative/null?→ Grade F
What does each grade mean in plain English?
The decision tree is precise; the labels are readable. Here is what each letter actually tells you about the human proof.
| Grade | What it means for you |
|---|---|
| Grade A | FDA-approved / proven in humans |
| Grade B | Real human trials, limited or historical |
| Grade C | Early / foreign human data only |
| Grade D | Animal studies only, unproven in humans |
| Grade F | Tested in humans and failed |
Does affiliate money influence the grades?
No, and this is the whole point of the site. Grades are assigned by the decision tree above, from the published evidence, full stop. A peptide we cannot monetize still gets the grade its evidence earns; a legal access route we can monetize never lifts a grade. If a famous, heavily-marketed peptide has only animal data, it gets a D on this page no matter how much traffic or revenue a higher grade would bring. That editorial firewall is why the grades are worth trusting.
The harder call is when human data exists but belongs to something else. The follistatin research peptide is the cleanest example on this site: the only credible human work delivered the gene by an AAV vector so muscle would express the protein continuously, which is gene therapy and not the vial anyone is selling. The tree above asks what evidence exists for the thing being sold, so the injectable is graded on its own record rather than borrowing a trial it was not in.
The firewall, stated plainly
Evidence grade answers one question only: how strong is the human proof? Whether we can earn a commission is a completely separate axis, handled on the cost & access page, and it never touches the grade.
How do you classify legal status?
Evidence grade tells you whether a peptide works. Legal status is a separate question: whether you can access it through a legal, supervised route right now. Every peptide gets exactly one of these five values.
- FDA-approved
- Prescription-legal via a licensed provider. The drug cleared the full FDA approval process for at least one indication.
- Compoundable (503A)
- Legal to obtain through a 503A compounding pharmacy or licensed telehealth, typically because of a prior FDA approval that keeps it on the compounding pathway.
- Under FDA review
- In regulatory limbo. The FDA's Pharmacy Compounding Advisory Committee votes July 23 to 24, 2026 on whether it can be legally compounded. No settled legal supervised route until then.
- Research-only
- No legal supervised US route. Sold only as "research use only," which is a gray-market status, not a legal way to use it in a person.
- Cosmetic/topical
- Legal as a topical cosmetic ingredient (e.g. copper-peptide serums), a different regulatory lane from injectable drugs.
What does the safety flag mean?
A third, independent signal: a green / amber / red dot for how well the human safety of a peptide is characterized. It is not a verdict on whether a peptide is “safe to take.” It flags how much we actually know.
- green
- Well-characterized human safety (FDA-approved or long clinical history).
- amber
- Limited human safety data, no major documented harms.
- red
- Documented serious harms, or disproven but still sold.
What a grade looks like when you check it against the registry
The rubric above turns on published human evidence, which invites a fair objection: how would you know? For the most-searched compound on this site, you can check us. Every human trial is supposed to be registered publicly before it enrols, so we queried the registry for BPC-157 on September 4, 2026. It returned 4 records, and 3 of them has a design capable of showing whether the compound works.
Not one has published a result. The oldest stopped reporting a few days after it was registered and still sits at status UNKNOWN. One is a single-arm, unmasked supplement study that reached the public registry after it had already finished. The only trial designed to answer the efficacy question is still recruiting, with an estimated primary completion of February 14, 2027. That is the arithmetic behind the D, and it is the reason a committee vote in July did not move it.
The count moved while this page was live, which is the honest way to show what a grade does and does not depend on. The same query returned 3 records on August 24, 2026 and 4 on September 4, 2026. The new one is NCT07803250, first posted September 3, 2026, sponsored by University of Arkansas. Its own registration carries the sentence the grade rests on: it has not yet been studied in formal human clinical trials
. That is the sponsor writing it on the registry, not our reading of the literature. The grade did not move and could not have: the registry lists the study as not yet recruiting, with no results filed and enrollment estimated to begin in January 1, 2027. A registered trial is a plan to produce evidence. It is not evidence, and nothing on this page moves until it reports.
Where that pending trial runs is part of the arithmetic too, so we state it rather than leaving it to the registry link. NCT07437547 lists a single study site, Peking University Shenzhen Hospital in Shenzhen, China, and its own oversight module records isFdaRegulatedDrug: false. That is the sponsor's attestation on its registration form, not a finding by the FDA and not a refusal by it. We flag it here because the grade is a claim about evidence, and a reader deciding how much weight to put on a result due in February 14, 2027 should know it is coming from one site abroad, in a study whose sponsor says it sits outside FDA drug regulation. It does not make the trial worse. It does mean the eventual result arrives without the FDA having overseen its conduct, which is a different thing from a US pivotal trial and should not be read as one. Record read at the ClinicalTrials.gov API on August 26, 2026, HTTP 200.
Why a recommendation cannot move a grade
Grades answer “does the human evidence support this?” Legal status answers “may a pharmacy make it?” They are different questions with different inputs, and neither one is an argument for the other. If the FDA adds BPC-157 to the 503A list tomorrow, the grade on this site does not move, because nothing about the published evidence would have changed. The registry footprint, sponsor by sponsor and date by date, is on the FDA status tracker.
How do the peptides we track fall across the grades?
This is the live distribution across all 46 peptides on the site, computed from our compound records. Tap any name for its full evidence review and sources.
See the same set in one sortable place on the full peptide table. Evidence grade and legal status are separate questions, and the second one moved in July 2026: we track every peptide vote the FDA advisory committee took, tally by tally, on the status tracker.