DoseProof

Testing guide

Understanding lab results

What each number on a GLP-1 lab report means, what a good result looks like, and what a result can't tell you.

Last reviewed 16 Nov 2026

What each test does, and doesn't

A lab report is only as useful as the tests behind it. No single test answers every question, so check which ones were run before reading the numbers.

LC-MS

Liquid chromatography–mass spectrometry

Separates the sample on a column, then turns the molecules into charged particles and weighs them. Software converts the readings into the molecule's mass.

Tells you

  • Whether the expected molecule is present (identity)
  • Whether a different peptide is there instead, by its mass
  • The masses of some impurities

Doesn't tell you

  • How much of the peptide there is, unless the lab runs a separate quantitative method
  • That the peptide is the main thing in the vial: a small amount of the right molecule still reads "identity confirmed"
  • Anything that doesn't ionise well, such as salts and many fillers
  • Molecules with exactly the same mass, such as some altered forms of the peptide
  • Bacteria, endotoxin or sterility

On the report: "Identity confirmed" plus the mass found, in daltons (Da)

HPLC

High-performance liquid chromatography, usually with a UV detector

Pumps the sample through a column that holds back each component for a different time. A UV detector draws each component as a peak as it comes out.

Tells you

  • Purity: the main peak's share of all peaks
  • Quantity in milligrams, when compared with a reference standard of known amount
  • Identity, but only weakly: a peak at the same time as a reference standard

Doesn't tell you

  • Proof of identity on its own: two different peptides can come out at the same time
  • What the impurity peaks actually are
  • Water, salts and other things that don't absorb UV light
  • Bacteria, endotoxin or sterility

On the report: Purity in %, quantity in mg, and a chromatogram

See a real HPLC example →

LAL test

Limulus amebocyte lysate (endotoxin test)

Mixes the sample with a reagent made from horseshoe crab blood, which clots or changes colour when bacterial endotoxin is present. The strength of the reaction gives the amount.

Tells you

  • How much endotoxin is present: the toxin from the cell walls of certain bacteria, which can cause fever and shock when injected

Doesn't tell you

  • Whether live bacteria or fungi are present
  • Toxins and other fever-causing substances that aren't endotoxin
  • Identity, purity or quantity
  • A reliable value if the sample interferes with the reagent; good labs check for this

On the report: Endotoxin units (EU) per mg or per vial, often "< x" when below the detection limit

Sterility test

Culture or rapid PCR

A culture test puts part of the sample in growth media for about 14 days and watches for growth. A rapid test uses PCR to look for bacterial DNA.

Tells you

  • Whether microorganisms were found in the portion tested

Doesn't tell you

  • That every vial in the batch is sterile: only part of a sample is tested
  • Endotoxin: a sterile sample can still contain toxin from bacteria killed earlier
  • With PCR: whether the bacteria found were alive, or organisms the test isn't designed to detect

On the report: Pass or fail ("no growth", "not detected")

Which test answers which question

TestRight molecule?How pure?How much?Endotoxin?Sterile?
LC-MSYesPartlyPartlyNoNo
HPLCPartlyYesYesNoNo
LAL testNoNoNoYesNo
Sterility testNoNoNoNoYes

"Partly" for LC-MS: some labs can estimate purity or amount by LC-MS, but most report it for identity only. "Partly" for HPLC identity: a matching retention time supports identity but doesn't prove it. A full picture of an injectable needs LC-MS and HPLC, plus endotoxin and sterility if you want to know it is safe to inject.

Identity: is it the right molecule?

Identity is usually tested with LC-MS, which weighs the molecule. The report gives the mass found, often after the software combines several charged forms of the peptide into one value. Compare it with the expected mass below. A small difference, around 1 to 2 daltons, can come from the instrument or the calculation used. A large difference means a different molecule.

PeptideFormulaExpected average mass
SemaglutideC187H291N45O59≈ 4113.6 Da
TirzepatideC225H348N48O68≈ 4813.5 Da
RetatrutideC221H342N46O68≈ 4731.4 Da

What an LC-MS identity result looks like

Illustrated LC-MS identity result for tirzepatide: chromatogram, mass spectrum with 3+ to 6+ charge states, and deconvoluted mass of about 4813 Da 1 · Chromatogram: the peptide comes off the column at one time peak at 6.8 min retention time (min) → 2 · Mass spectrum of that peak: the same molecule carrying 3, 4, 5 or 6 charges 803.26+ 963.75+ 1204.44+ 1605.53+ 800 1000 1200 1400 1600 m/z (mass ÷ charge) → 3 · Deconvoluted mass: software combines the charge states into one molecular weight semaglutide would be here (4113.6) retatrutide would be here (4731.4) found4813.4 Da= tirzepatide 4000 4250 4500 4750 5000 molecular weight (Da) →
Illustration built from tirzepatide's calculated masses, not a real measurement. Real spectra show each ion as a small cluster of isotope peaks and some background noise. A real report may show any of these three views; the deconvoluted mass is the one to compare with the table above.

A real spectrum from lab

Real ESI-TOF mass spectrum of a peptide of about 2174 Da, with charge states at 435.8, 544.5, 725.7 and 1088.1, and below it an MS/MS fragmentation spectrum of the 544.3 ion
Real data from an LC-MS (ESI-TOF), sample details removed. This peptide is not a GLP-1: it weighs about 2174 Da. The reading method is the same for any peptide.

Top panel: full mass spectrum

One molecule appears as several peaks because it picks up different numbers of charges: 1088.1 (2+), 725.7 (3+), 544.5 (4+) and 435.8 (5+).

The small peaks next to each main peak are isotopes. Their spacing gives the charge: 544.3, 544.5, 544.8 are 0.25 apart, so that ion carries 4 charges.

Mass = (544.5 − 1.007) × 4 ≈ 2174 Da. Every charge state gives the same answer, which is a good sign that it is one molecule.

Bottom panel: MS/MS fragmentation

The 4+ ion was isolated and broken apart. The fragments form a fingerprint of the peptide's amino-acid sequence.

This goes further than mass alone: it can tell apart two molecules with the same weight. Most consumer lab reports show only the top panel or the deconvoluted mass.

HPLC can also suggest identity by comparing retention time with a reference standard, but two different peptides can come out at similar times. LC-MS is the stronger proof.

Purity: how much of it is the peptide?

HPLC separates the sample into its components and draws each as a peak. Purity is the main peak's share of all the peaks, as a percentage. Many sellers claim 98% or more.

Example HPLC chromatogram: one large main peak and three small impurity peaks Main peak: the peptide impurity impurity impurity Retention time (minutes) → Signal Purity = main peak area ÷ all peak areas
Illustration, not a real result. A clean sample shows one dominant peak; every extra peak is something else in the vial.

Quantity: is the amount right?

Quantity, sometimes called content or assay, is how many milligrams of peptide the vial actually holds. Compare it with the label: 9.86 mg in a vial labelled 10 mg is 98.6% of the label claim. Licensed medicines are typically held to roughly 90–110% of the label; grey-market products have no required range.

How the lab measures it

Quantity is normally measured by HPLC against a reference standard: a sample of the same peptide whose amount and purity are already known.

  1. Dissolve the whole vial

    The lab dissolves the contents in an exact volume of liquid.

  2. Run the reference standard

    Known amounts of the standard are run first, usually at several concentrations, to see how big a peak each amount produces. This is the calibration.

  3. Run your sample

    The sample is run under the same conditions. Its main peak should appear at the same retention time as the standard's.

  4. Compare peak sizes

    The size of the sample's peak, read against the calibration, gives the concentration. Multiplied by the volume, that is the milligrams in your vial.

Two chromatograms stacked: the reference standard and the sample, with main peaks at the same retention time Reference standard (known amount) Your sample peak area = known mg area vs. standard → mg in vial Main peaks line up at the same retention time time →
Illustration, not a real result. The sample's main peak should sit at the same time as the standard's, and its size, compared with the standard, gives the amount.

What a good chromatogram shows

Look forWhy it matters
Sample name or report number printed on the chromatogramTies the picture to your report, not someone else's.
A time axis and signal axis with unitsLets you see where peaks fall and compare with other reports.
One sharp, symmetrical main peakA clean sample. A broad, split or shouldered peak can hide another substance.
The main peak's retention time, matching the standard'sSupports that the main peak is the right peptide.
A flat baseline, shown in full from start to endA cropped picture can hide extra peaks.
A peak table: retention time, area and area % for each peakShows how the purity figure was calculated.

Is the reference standard's chromatogram included?

Often not. Many lab reports for consumers show only the sample's chromatogram and the final numbers. The lab keeps the standard's data and calibration in its own records.

That isn't a problem in itself, but it means you can't check two things yourself: that your main peak matches the standard's retention time, and that the quantity was calibrated against a real standard of the same peptide. If it matters, ask the lab:

  • Which reference standard was used: its source, lot and purity.
  • Whether it is the same peptide as your sample.
  • How many calibration points were run.
  • Whether they can send the standard's chromatogram.

A lab that reports milligrams but can't name the standard it used has given you an estimate, not a measurement.

A real HPLC example: reference standard vs sample

Below are two real HPLC runs on the same instrument with the same method: a reference standard and a sample of the same peptide. Names, file paths and dates are removed; the peptide is not a GLP-1, but the reading is the same for any peptide. Detection is UV at 220 nm on a C18 column, over a 40-minute gradient.

Two real HPLC chromatograms stacked. Reference: one main peak at 11.217 minutes. Sample: main peak at 11.146 minutes with two small impurity peaks at 10.798 and 13.064 minutes. Both show a solvent front at about 2 minutes.
Real data, annotated. The dashed line marks the reference's main peak. The spikes at about 2 minutes in both runs are the solvent / injection front, not peptide, so they are left out of the purity calculation.

Reference · 5.0 µL injected

PeakTime (min)Area (mAU·s)Area %
111.2173171.92100.00

Sample · 10.0 µL injected

PeakTime (min)Area (mAU·s)Area %
110.79820.360.32
211.1466273.6599.29
313.06424.750.39

What the two runs tell you

ReferenceSampleWhat it means
Main peak time11.217 min11.146 min0.071 min apart (0.6 %): inside accepted tolerance
Purity100 %99.29 %Two small impurities either side of the sample's main peak
Injected volume5.0 µL10.0 µLSample injected at twice the volume, so its peak is about twice as tall
Area per µL injected634.4627.4Sample concentration is 98.9 % of the reference's
  • Amount: peak area divided by the injected volume puts both runs on the same scale. The sample gives 98.9 % of the reference's signal per µL, so if the reference concentration is known, the sample's concentration is 98.9 % of it.
  • Purity: both runs are integrated the same way, with the solvent front excluded. Integration rules matter: when the same reference was integrated with the solvent front included, the software reported 74.7 % instead of 100 %. Always check that every run was integrated by the same rules.

How far apart can retention times be?

Retention time always moves a little between runs: with the mobile phase, column temperature, column age, flow rate and the injected volume. Published identity criteria for LC methods set these limits:

CriterionLimitThis example
Sample vs standard retention time (SANCO guidance)± 0.2 min0.071 min: within
Relative retention time vs an internal standard (Decision 2002/657/EC)± 2.5 %No internal standard used
Relative retention time (Implementing Regulation (EU) 2021/808)± 1 %0.6 %: within
Peak retained at least twice the dead time (k ≥ 1)k ≥ 111.2 min vs ~2 min front: met
Peak shape matches the standard'sSame shapeBoth sharp and symmetrical: met

A match is necessary, but it isn't proof

If the retention time and peak shape match, the right peptide may be present, but more confirmation is needed. If they don't match, the peak can't be that peptide. Retention time can rule a compound out; it can't rule it in.

Why a different peptide can land in the same spot

How an HPLC-only identity check can be fooled

Each of these passes a report that shows only retention time and purity. Knowing them tells you what to ask for.

What's in the vialWhy HPLC alone passes itWhat catches it
A cheaper peptide with similar column behaviourComes out inside the retention-time window as one clean peakMass by LC-MS; co-injection with a certified standard
A crude or wrong-sequence version of the right peptideNear-copies come out at almost the same time, or togetherHigh-resolution MS, MS/MS sequencing, a second column of different type
A blend of the right peptide and a cheaper filler peptideBoth merge into one peak, and purity stays highMass spectrum across the peak; a second column
Anything, compared with a "reference" that isn't independentA sample always matches its reference if they are the same materialA reference standard from an independent, certified source, with its own lot number
Peaks left out of the calculationPeaks that aren't integrated don't appear in the purity figureThe full chromatogram and peak table; the same integration rules for every run
A different drug altogether in the right packagingRetention time and purity alone don't name the moleculeLC-MS: in Germany, high-resolution LC-MS and NMR showed that fake GLP-1 pens contained an insulin instead

What does confirm identity

Verdict for this example: the retention times match within every tolerance above and the peak shapes match. That is consistent with the same peptide but does not confirm it: no co-injection was run and no mass spectrum is included. To confirm identity, add an LC-MS mass result and a co-injection with a certified reference standard.

Endotoxin and sterility: is it safe to inject?

Endotoxin is a toxin from bacterial cell walls that can cause fever and serious reactions when injected. It is measured with the LAL test in endotoxin units (EU), per milligram or per vial. Lower is better; the lab should state the limit it compares against. Filtering a mixed solution removes bacteria but not endotoxin.

Sterility is reported as pass or fail. A culture test takes about two weeks. Faster tests detect bacterial DNA instead. A pass on the tested sample can't guarantee every vial in the batch is sterile.

Reading example results

LabelIdentity (LC-MS)PurityQuantityWhat it means
Tirzepatide 10 mgMolecular weight of tirzepatide: 4813.5 DaFound in vial: 4813.4 Da→ Confirmed: tirzepatide99.2%9.86 mgMatches label Right molecule, clean, right amount.
Tirzepatide 10 mgMolecular weight of tirzepatide: 4813.5 DaFound in vial: 4813.6 Da→ Confirmed: tirzepatide98.7%6.10 mgUnderdosed 61% of the label. Every measured dose is too low.
Semaglutide 5 mgMolecular weight of semaglutide: 4113.6 DaFound in vial: 4113.5 Da→ Confirmed: semaglutide82.4%4.90 mgImpure Right molecule, but almost a fifth is something else.
Tirzepatide 10 mgMolecular weight of tirzepatide: 4813.5 DaFound in vial: 4113.6 Da→ Not confirmed: the mass matches semaglutide (4113.6 Da)98.9%—Wrong molecule Sold as tirzepatide, but it is semaglutide. Every dose is a different drug at a different strength.
Retatrutide 10 mgMolecular weight of retatrutide: 4731.4 DaFound in vial: no peptide signal→ Not confirmed: no peptide detected—0 mgNo peptide The vial doesn't contain what it claims.

Example results for illustration, not real reports. A found mass within about 1–2 Da of the expected value counts as a match. When it doesn't match, compare it with the other GLP-1 masses in the identity table above: a fake can contain a different peptide altogether.

Results apply to the sample tested

A good result covers that one vial, on that day. A new batch needs a new test. If your result shows a problem, see what to do next. This page is for education, not medical advice.