The evidence register
Start with the goal. Follow the evidence.
Peptideworth grades every compound it tracks by how far the published human evidence actually goes — and says plainly when that is nowhere.
Of the 13 compounds in this register, 6 have randomized human trials behind them, 5 have early human data only, and 2 have never been published in a person at all. 7 of the best evidenced cannot legally be bought.
Last checked August 2026
- Compounds graded
- 13
- Papers read
- 79
- Companies audited
- 94
- Research articles
- 27
Four steps, weakest to strongest
We show the step the published evidence reaches — not the step the seller implies.
- Anecdotal0
User reports and forum consensus. No controlled data.
- Animal only2
Rodent or cell studies. Nothing published in humans.
- Early human5
Small, open-label, or phase 1/2 trials. Promising, unproven.
- Randomized trials6
Randomized controlled trials in humans, at scale.
Start here
What are you trying to fix?
6 goals, each ranking its compounds by how far the published human evidence actually goes. The grade beside each one is the strongest step any compound on that page reaches — not an average, and not a promise.
Weight loss
Four peptides have real weight-loss trials behind them. The rest have animal data or nothing. Here is what each one did, and what happens when you stop.
Randomized trials
4 of 8 at this tier · 9 papers
Weight-loss plateau
The weight curve flattening is a trial finding, not a failure. What STEP 5, SURMOUNT-4, STEP UP and SURMOUNT-5 measured about what happens next.
Randomized trials
3 of 5 at this tier · 11 papers
Muscle loss on a GLP-1
About a quarter of the weight lost on an incretin drug is lean mass. Here is what the trials measured, and which compounds have evidence for holding onto it.
Randomized trials
3 of 6 at this tier · 9 papers
Visceral fat
One peptide is FDA-approved specifically to reduce visceral fat — in one narrow population. Here is what it did, and what the GLP-1 drugs did alongside it.
Randomized trials
3 of 6 at this tier · 8 papers
Injury recovery
BPC-157, TB-500, GHK-Cu: the animal data is genuinely good and the human data barely exists. What is registered, what has reported, and what has not.
Early human
1 of 5 at this tier · 7 papers
Sleep
One peptide has a randomized sleep-outcome trial behind it, and it is not one of the peptides sold for sleep. What the trials measured, and what they did not.
Randomized trials
1 of 5 at this tier · 5 papers
The evidence and the availability do not line up.
Two lists fall straight out of the register, and they point in opposite directions. Neither is a recommendation — one is a supply fact, the other is an evidence fact.
Well evidenced, not available
7- SermorelinCompounding restricted
- SurvodutideResearch use only
- CagrilintideResearch use only
- GHK-CuCompounding restricted
- GlutathioneCompounding restricted
- NAD+Compounding restricted
- RetatrutideResearch use only
Real trial data, no legal route to buy them for use in a person. If a seller offers you one of these, that is the fact worth knowing first.
The registry
Every compound we have reviewed
One page per molecule, whatever its evidence tier — including the ones where the honest answer is animal data only.
Ordered by how far the published human evidence goes, strongest first — then alphabetically. Nothing else moves a row.
| Compound | Strongest evidence | Availability | Papers |
|---|---|---|---|
| Bremelanotide (PT-141) | Randomized trials | FDA-approved | 7 |
| Semaglutide | Randomized trials | FDA-approved | 7 |
| Sermorelin | Randomized trials | Compounding restricted | 5 |
| Survodutide | Randomized trials | Research use only | 6 |
| Tesamorelin | Randomized trials | FDA-approved | 9 |
| Tirzepatide | Randomized trials | FDA-approved | 9 |
| Cagrilintide | Early human | Research use only | 6 |
| GHK-Cu | Early human | Compounding restricted | 5 |
| Glutathione | Early human | Compounding restricted | 5 |
| NAD+ | Early human | Compounding restricted | 4 |
| Retatrutide | Early human | Research use only | 8 |
| BPC-157 | Animal only | Compounding restricted | 3 |
| TB-500 | Animal only | Research use only | 5 |
The companies
105 of 379 product claims match the trial behind them.
We read what 94 companies say about every peptide they sell and check it against the published record. A claim that outruns its evidence is the single most common thing on a peptide storefront.
- Publishes a testing regime18
- Names the pharmacy that fills it38
- Publishes a readable certificate of analysis1
The research
What the evidence actually shows
- Compound Classes
- 11
- Sourcing & Compounding
- 3
- Dosing & Effects
- 3
- Reading the Evidence
- 10
Amylin Analogs: The Next Class After Incretins
An amylin analog has been FDA-approved since 2005. What changed is not the hormone — it is how long the molecule lasts. The trials, in order.
ReadCertificates of Analysis: What They Prove
A COA is a claim about a lot, not a guarantee about a vial. What each test answers, what a purity figure cannot say, and the gap the document never closes.
ReadCompounded Is Not the Trial Drug
A trial tests one defined article: a molecule, a form, a concentration, a container. Compounding changes some of those, and the evidence travels that far.
Read503A vs 503B: What Is Actually in the Vial
One scheme is inspected against manufacturing standards; the other is not. What each may legally start from, and what the inspection record shows in practice.
ReadGrowth-Hormone Secretagogues: What the Evidence Actually Shows
These compounds reliably raise IGF-1. That is the surrogate. When trials measured function, fracture recovery or disease progression, the results changed.
ReadHealing Peptides and the Gap Between Anecdote and Trial
BPC-157 and TB-500 have thirty years of animal work and a thin human record. Here is exactly what is registered, what is published, and what neither shows.
ReadTools
Calculators, not dosing advice
Each one converts numbers you already have — a vial's printed strength, how much water you mixed in, a dose you and a clinician have already chosen — into syringe units. None of them suggest a dose, a starting point, or a schedule.
- Reconstitution calculatorConvert a peptide vial's strength and your bacteriostatic water volume into insulin-syringe units for a dose you supply. Arithmetic only — not dosing advice.
- Retatrutide calculatorConvert a retatrutide vial and your bacteriostatic water into syringe units for a dose you set. Read the supply and approval facts first — they matter more.
- Microdosing split calculatorDivide a weekly total you have already decided on into smaller injections and see the syringe units for each. Your numbers only — no chart of amounts to copy.
- Unit converterConvert one quantity between micrograms, milligrams, milliliters and U-100 syringe units at a concentration you supply. Arithmetic only — never dosing advice.
- Bacteriostatic water calculatorWork out how much bacteriostatic water to add to reach a concentration you have chosen, and what one syringe unit is worth once you get there.
How we grade
Grant Delaney
Research Editor
- Graded, not asserted
- Every compound carries the evidence tier its published trials support — and keeps it when that tier is inconvenient.
- Numbers from the paper
- Every figure is quoted from the trial it cites, with the PMID to check it. No figure on this site was written from memory.
- Weak evidence stays weak
- We say when the answer is animal data, or nothing. On this register that is 2 of 13 compounds, and we do not round it up.
Read how we research and rank in our editorial policy, and how we stay independent on our about page.