What does HPLC actually measure?
High-performance liquid chromatography pushes a dissolved sample through a column packed with tiny particles under high pressure. In the reversed-phase version used for peptides, the particles are coated with hydrophobic chains and the solvent gradually becomes more organic over the run.
Molecules that cling to the particles come off later; molecules that prefer the solvent come off earlier. A detector at the column outlet, usually ultraviolet absorbance at 210 to 220 nm where the peptide bond absorbs, draws a trace of signal against time. Each substance that separates cleanly appears as a peak.
Purity is then a piece of arithmetic. Software integrates the area under each peak. The area of the retatrutide peak divided by the sum of all peak areas, multiplied by 100, is the purity in percent. A trace with one large peak and four tiny ones might integrate to 98.6 %.
A trace with a large peak and a broad shoulder might integrate to 94 %. The number is only as good as the method: a method that fails to separate an impurity from the parent peptide will hide it inside the main peak and report a purity that is too high.
Peptide chemists validate a stability-indicating method for exactly this reason. In a published example for a synthetic decapeptide, the authors screened buffers, organic modifiers and columns until six known related impurities separated from the parent peak, then stressed the sample with heat, acid, base and oxidation to confirm the method could see degradation products[1]. Retatrutide is a lipidated peptide of roughly forty residues with a mass near 4730 Da[2], which makes the separation harder, not easier.
Key numbers
What are the small peaks?
Synthetic peptides are built one amino acid at a time on a resin. Each coupling step is efficient but not perfect. Over roughly forty steps, a small fraction of chains will miss a residue (deletion sequences), stop growing early (truncated sequences), retain a protecting group that should have been removed, or pick up an oxidised methionine or a racemised residue.
Each of these has a slightly different hydrophobicity and therefore a slightly different retention time, which is why they show up as separate small peaks near the main one[1].
A deletion sequence of retatrutide is still a peptide that looks almost like retatrutide, will co-elute closely, and can show similar mass. This is why identity by mass spectrometry is reported separately from purity by HPLC: the mass spectrometer can distinguish a chain that is one residue short, while the UV detector cannot.
A COA that reports both is telling you the main peak has the right mass and that the wrong-mass material is small. Read how to read a retatrutide COA for the line-by-line version.
The peaks you should worry about are the ones that are not there. If a trace shows one perfectly symmetrical peak and a flat baseline with no noise, it may be a stock image or a trace from a reference standard rather than the lot in question.
Real chromatograms have baseline noise, a solvent front at the start, and a few small peaks. A vendor that shows the same trace across several products, or the same trace across several months of lots, is showing decoration.
The measured semaglutide purity was significantly low, ranging between 7.7 % and 14.37 % and deviating from the 99 % claimed on product labels.
Figure 1
What a purity figure cannot tell you
Purity is a ratio between what is peptide and what is not, among the things the detector can see. It cannot tell you how many milligrams are in the vial, because the vial was filled by weight of a powder that also contains water and counter-ion, and because filling error is a separate process from synthesis.
The 2026 analysis of retatrutide products in Australia found that three vials, all genuine retatrutide, contained 51 %, 165 % and 190 % of the labelled 10 mg[2]. Purity would not have flagged any of them.
Purity cannot tell you about endotoxin. Lipopolysaccharide fragments do not absorb at 214 nm in a way that produces a peak and they are present at microgram levels that UV detection does not resolve. The LAL test is a different assay with its own variables and its own interference problems in protein solutions[3].
A 2024 test-purchase study of GLP-1 peptide products sold online found endotoxin in every sample, between 2.16 and 8.95 EU per milligram, while measured purity was 7.7 % to 14.4 % against a 99 % label[4]. Both failures on the same vials. Only one of them would show on an HPLC trace.
Purity cannot tell you about sterility, about heavy metals (the Australian analysis detected lead at 0.7 mg/kg, below toxicological thresholds but present[2]), or about residual solvents.
Each is a separate line, and the honest answer on many research COAs is that these lines are absent. The point is not that every research lot needs every test. The point is that a 99 % figure is one test, and a listing that presents it as a summary of quality is overstating what was measured.
How to read a chromatogram in thirty seconds
- Find the main peak. It should be tall, narrow and roughly symmetrical, with a retention time stated on the report.
- Look at its neighbours. Small peaks just before and just after are normal synthesis-related impurities. A shoulder on the main peak is a co-eluting impurity the method did not fully resolve.
- Check the integration table. Each peak should have a retention time, an area and a percentage. The percentages should add to 100 and the main peak percentage should equal the stated purity.
- Check the baseline. Real traces have noise. A perfectly flat baseline suggests a smoothed or synthetic image.
- Match the header. Sample name, lot number, date and laboratory must match the certificate and the vial.
If the listing shows a purity number but no trace and no integration table, treat the number as a claim. Ask for the report. A vendor that runs the test can send it in minutes. That is the difference we describe in grey market versus research vendors, and it is the first thing to check in the buying guide.
Why purity matters for a research compound that is still in trials
Retatrutide is investigational. The phase 3 TRIUMPH trials in obesity finished enrolment and treatment in 2026 with 2 335, 1 152 and 1 946 participants in the three master-protocol studies[5][6][7], and the phase 3 type 2 diabetes trial TRANSCEND-T2D-1 has been published[8].
Those trials used GMP material with full release testing. Material sold as a research chemical has no such release, which is why the certificate carries the whole burden of proof.
For a laboratory, purity determines whether an observed effect in an assay belongs to retatrutide or to an impurity, and it determines how a stock solution is calculated. A 90 % lot treated as 100 % pure is a 10 % error before anything else goes wrong.
Combined with a content error of the size seen in the Australian study, the compounded error can approach a factor of two in either direction. For anyone handling the material in a research setting, the practical steps are in reconstitution and storage.
None of this makes retatrutide an approved medicine, and it is not one. It is not approved by the FDA or the EMA, cannot be prescribed, and is not on any compounding list[9]. The legal situation is described in is retatrutide legal.
Questions readers ask
Is 99 % HPLC purity good for retatrutide?
Yes, 98 % to 99 % is typical of a clean synthetic lot. But purity is a ratio of peak areas; it does not tell you how many milligrams are in the vial or whether endotoxin is present. Ask for net content and endotoxin as separate lines.
Can a fake product show 99 % purity?
A product can show a purity figure copied from elsewhere, and a chromatogram can be a stock image. A trace tied to a lot number from a named third-party laboratory with a report number is the only version worth trusting. In a 2024 test-purchase study, products labelled 99 % measured 7.7 % to 14.4 % (DOI 10.2196/65440).
What is the difference between HPLC purity and mass spec identity?
HPLC purity tells you what fraction of the detectable material sits in the main peak. Mass spectrometry tells you the main peak has the right mass, near 4730 Da for retatrutide. You need both.
Does purity change over time?
It can. Peptides degrade with heat, moisture and light, and oxidation or deamidation shows up as new peaks. That is why the report date matters and why storage conditions matter after the vial reaches you.
