Peptides, Ranked by Evidence: What Has a Phase 3 Trial and What Has a Forum Post
Peptides sold through the same websites sit at completely different evidence levels. Semaglutide and tirzepatide have multiple phase 3 trials and approved labels. Tesamorelin has an approved indication. Sermorelin was approved and withdrawn for commercial reasons. BPC-157 and TB-500 have no human efficacy trials at all — and BPC-157 was placed on the FDA's 503A Category 2 bulks list, meaning it is not permitted for compounding.
The tiers
| Peptide | Evidence tier | What exists | Legal status for compounding |
|---|---|---|---|
| Semaglutide | Tier 1 | Multiple phase 3 trials; outcomes trial (SELECT) | Approved product; 503A patient-specific only |
| Tirzepatide | Tier 1 | Multiple phase 3 trials; head-to-head win | Approved product; 503A patient-specific only |
| Liraglutide | Tier 1 | Phase 3 trials; two FDA-approved generics exist | Approved; generics available |
| Tesamorelin | Tier 2 | Approved for a specific indication | Approved product |
| Sermorelin | Tier 3 | Was approved, withdrawn for commercial reasons | Compounded; not a current approval |
| Ipamorelin | Tier 4 | No approved indication; limited human data | Category 2 bulks list |
| BPC-157 | Tier 4 | No human efficacy trials | Category 2 — not permitted |
| TB-500 | Tier 4 | No human efficacy trials | Not permitted |
Tiers describe the weight of evidence, not safety or effectiveness for any individual.
Why does a peptide with no trials get sold at all?
Three routes, and they are not equivalent.
- "Research use only" labelling. Sold as a chemical, not a drug, with a disclaimer that it is not for human use. The disclaimer is the legal shield; it is also an admission.
- Compounding. A 503A pharmacy preparing a substance for an individual patient. This is the route Category 2 closes.
- Grey-market import. Outside the regulated system entirely.
None of these produces evidence. A product can be widely sold, widely reviewed and entirely untested — availability is not evidence, and a large body of user reports is not a trial.
Indicative count of registered phase 2/3 trials with a published efficacy endpoint in the relevant indication, compiled 4 August 2026. Zero means zero.
What to ask before buying any peptide
- Is there a human trial with a published efficacy endpoint? Not a mechanism study, not an animal model, not a case series.
- Is it on the Category 2 bulks list? If so, no legitimate 503A pharmacy should be compounding it.
- Which pharmacy fills it, and under 503A or 503B? A named pharmacy can be checked against its state board.
- Does the label say "not for human use"? If so, the seller has already told you what it is.
Tier 4 does not mean "harmful". It means nobody has run the study that would tell you, and the absence of evidence of harm is not evidence of safety.
What "no human trials" actually means
It does not mean nothing has been published. BPC-157 has a substantial animal literature and a mechanistic story about angiogenesis and tissue repair that is genuinely interesting. What it does not have is a randomised controlled trial in humans with a published efficacy endpoint.
That distinction is the whole argument. An animal model tells you a compound does something in a rat. A mechanism study tells you a plausible pathway exists. Neither tells you the dose, the effect size, the adverse-event profile or the interactions in a person — which is what a trial is for, and why approval requires one.
| Evidence type | Can establish | Cannot establish |
|---|---|---|
| Mechanism study | A plausible pathway | That it works in a person |
| Animal model | An effect in that species | Human dose, safety or effect size |
| Case series | That something was observed | That the drug caused it |
| Phase 3 RCT | Effect size against a control | Rare long-term harms |
| Outcomes trial | That events are reduced | Results in an unstudied population |
Peptides in Tier 4 have the first two rows. The approved GLP-1s have all five.
Why the GLP-1s are a different category entirely
Semaglutide and tirzepatide are not just better-studied peptides. They have completed the full sequence: phase 3 efficacy trials, approved labels with dosing and contraindications, post-marketing surveillance, and in semaglutide's case a cardiovascular outcomes trial reporting a reduction in major adverse cardiovascular events.
A site listing them alongside BPC-157 under one heading — "peptides" — makes a chemical classification do work that an evidence classification should be doing. They are the same kind of molecule and utterly different products.
Where sermorelin sits, and why it confuses people
Sermorelin was an approved product. It was withdrawn from the market for commercial reasons rather than a safety finding, which is why it occupies an odd position: it once cleared an approval bar that BPC-157 has never approached, and it is not a current FDA-approved product today.
Providers sometimes cite the historic approval as though it were current. It is a real fact about the past and not a statement about what is in the vial now, which is a compounded preparation.
Four questions that sort any peptide in under a minute
- Name the trial. If a seller cannot name a human trial with an efficacy endpoint, there is not one.
- Check the bulks list. Category 2 means a legitimate 503A pharmacy cannot compound it.
- Read the label. "Research use only" and "not for human use" are the seller telling you what the product is.
- Ask who fills it. A named pharmacy can be checked against its state board and the FDA warning-letter database. An unnamed one cannot.
Why the tiering matters more than a safety verdict
It would be easier to write "Tier 4 peptides are dangerous". That is not what the evidence supports, and saying it would be the same error in the opposite direction.
The accurate statement is narrower and more useful: nobody has run the study that would tell you either way. An absence of published harm is not a finding of safety, and a large body of enthusiastic user reports is not a substitute for a control arm. Tier 4 is a statement about what is known, not a verdict about the molecule.
The regulatory question is separate from the evidence question
These two get merged constantly, and they move independently. A substance can be well-evidenced and unavailable, or poorly evidenced and freely sold.
| Strong evidence | Weak or no evidence | |
|---|---|---|
| Legally available | Semaglutide, tirzepatide, liraglutide, tesamorelin | Sermorelin (compounded, not a current approval) |
| Restricted or not permitted | — | BPC-157 (Category 2), ipamorelin |
A product's presence on a website tells you nothing about which cell it occupies.
What this page does not settle
- Whether Tier 4 peptides work. No human efficacy trial exists to answer it.
- Whether they are safe. Same reason. The FDA's Category 2 designation reflects identified safety concerns, not a completed safety assessment.
- Whether the evidence will change. Trials could be run. Until they are, the tier stands.
If you are considering any peptide in this category, the highest-value thing you can do is ask the seller to name a human trial with a published efficacy endpoint, and see what comes back.