What a Retatrutide COA Should Contain: 8 Checks
A retatrutide COA should contain 8 things: lot, identity, method, mass, purity, endotoxin units, test date, lab. Only 3 check against an outside record.
A retatrutide COA should contain eight things: a lot code, the compound identity written as retatrutide with its development code, the name of the lab that ran the sample, the date it was run, the method behind the purity figure, the purity result with its basis stated, a mass confirmation, and an endotoxin result carrying its units. Three of those eight can be checked against something outside the certificate. The other five you either accept as written or pay a lab to re-run.
That split is the useful way to read the document. Sorting the fields by whether an independent check exists tells you where scrutiny actually buys you something.
What eight things should a retatrutide COA contain?
Run the certificate in this order. The first three rows resolve against an external reference; rows four through eight resolve only against the certificate itself.
| # | Field | What it should say | Independent check |
|---|---|---|---|
| 1 | Lot / batch code | A code on the certificate that also appears on the vial label | The vial you received |
| 2 | Compound identity | “Retatrutide”, plus the development code LY3437943 carried in the phase 2 obesity trial record | The published trial record |
| 3 | Testing lab | A named entity with an address or registration number you can look up | The lab’s own listing |
| 4 | Date of analysis | The date the sample was injected, not the date the page was built | None |
| 5 | Purity method | The method named, normally reverse-phase HPLC, with the detection wavelength | None |
| 6 | Purity result | A percentage with its basis stated as percent of total peak area | None |
| 7 | Mass confirmation | Observed mass against theoretical mass, by mass spectrometry | None without re-testing |
| 8 | Endotoxin | A figure with units, EU/mg or EU/mL, plus the concentration if quoted per mL | None without re-testing |
Caption: The eight-field read on a retatrutide COA, ordered by whether an outside reference exists for the field. The development code in row 2 is sourced to the phase 2 obesity trial report cited below; every other row describes document structure rather than a measured value.
Rows 1 and 2 cost you nothing to check and reject the largest share of bad documents. A certificate with no lot code cannot be tied to any vial, and a certificate naming a house abbreviation instead of the compound has not made an identity claim at all. The remaining fields are where most listings concentrate their marketing, and they are the ones you have the least leverage over.
Which COA fields can you verify from outside the document?
Three. The lot code against the vial label, the compound name against the published record, and the lab against whatever registry it claims membership in.
The identity check is the one people skip. Retatrutide is a single peptide with agonist activity at the glucose-dependent insulinotropic polypeptide, GLP-1 and glucagon receptors, studied at once-weekly maintenance doses of 0.5, 4, 8 and 12 mg in a phase 2 trial across 42 US centres that randomised 281 participants under registration NCT04867785. A COA naming “RETA” or “GLP-3” and nothing else gives you no way to line the vial up against that record.
The lab check is cheap and often decisive. A certificate signed by an unnamed “independent third-party laboratory” has removed the only field that could be falsified by a search. Where the lab is named, you can at minimum confirm it exists and runs peptide work.
Everything below row 3 is a closed loop. The purity percentage, the mass result and the endotoxin figure are all assertions the certificate makes about itself, which is why purity and potency claims diverge on listings that quote both without a quantitative assay behind either.
Why does the lot code matter more than the purity percentage?
Because a certificate covers one production run, and any research series of meaningful length spans several. The phase 2 obesity trial report ran its top arm at 12 mg once weekly for 48 weeks in a population of 338 adults. Translating that schedule into material: 12 mg × 48 weeks = 576 mg for one participant-equivalent.
| Vial format | mg per vial | Vials for 576 mg | Price per vial (USD) | Total (USD) |
|---|---|---|---|---|
| 10 mg | 10 | 57.6 | 100 | 5,760 |
| 20 mg | 20 | 28.8 | 195 | 5,616 |
Caption: Material required to cover a 12 mg once-weekly, 48-week schedule at the two listed vial formats. The 576 mg figure is 12 mg × 48 weeks from the phase 2 obesity trial report; vial sizes and prices are from the HEEZ catalog as of September 2026. Vial count is 576 ÷ 10 = 57.6 and 576 ÷ 20 = 28.8; totals are 57.6 × $100 = $5,760 and 28.8 × $195 = $5,616.
Fifty-seven vials do not come off one lot on a research listing. The lot on your fourth reorder will differ from the lot on your first, and a site-wide COA page carrying a single undated percentage cannot tell you which one you have. That is the argument for treating row 1 as load-bearing and row 6 as informational.
Vial format also decides how many lot boundaries a series crosses, and catalogues differ on the sizes they publish: HEEZ Research lists retatrutide in 10 mg and 20 mg formats as of September 2026, which is the pair the table above uses.
The wider field list, including the sterility and appearance lines that sit alongside these eight, is covered in the nine-field certificate read on this site.
What does an endotoxin figure without units actually tell you?
Nothing. “Endotoxin tested” printed as a badge is a claim that a test happened, not a result, and a bare numeral without EU/mg or EU/mL is unreadable in either direction.
Units change the figure by a factor equal to the concentration. Reconstitute a 10 mg vial with 2 mL of bacteriostatic water and the concentration is 10 mg ÷ 2 mL = 5 mg/mL. Converting a per-millilitre limit to a per-milligram one at that concentration is EU/mg = EU/mL ÷ 5, so the same numeral read as EU/mL is five times stricter than the same numeral read as EU/mg. A certificate quoting EU/mL without naming the concentration the lab tested at has not given you a convertible number.
A dated result also decays in relevance. The endotoxin and appearance lines describe a sample at the moment of testing, and nothing on the certificate speaks to transit or to how the lyophilised material was held before reconstitution. Research-listing material is supplied for laboratory use only and is not the material used in the trials cited here.
How do you tell a lot-specific COA from a template?
By whether the document changes between lots. A template is filled once and re-uploaded; a real certificate carries fields that move.
The three that move are the lot code, the analysis date and the purity figure’s trailing digits. Batch-to-batch variation in synthesised peptide purity is ordinary, so a vendor whose purity result shifts in the trailing decimals from one lot to the next is showing you more than one quoting a single fixed number across every batch. A page carrying the identical percentage across every lot it has ever listed is either publishing a specification limit rather than a measured result, or is not re-testing.
Chromatogram detail is the harder tell. Where a certificate reproduces the HPLC trace, the retention time of the main peak and the integrated peak areas are run-specific. If two lot-numbered certificates from the same vendor carry identical retention times and identical peak-area figures, they describe a single injection whatever the lot codes at the top say, and comparing two listings field by field is how that surfaces.
A certificate is also silent on the clinical record. The published evidence on retatrutide runs to three randomised controlled trials covering 640 patients, 510 of whom received the compound, in a 2024 systematic review and meta-analysis; a COA establishes what a lot contains and makes no claim connecting that lot to those results.