Purity percentage: what it proves

The number is a proportion of what was detected

A purity percentage from HPLC is relative peak area: the main peak as a share of the total area integrated by that method under those conditions. The sample is pushed through a column that separates molecules by how strongly they interact with it, a UV detector records what comes off, and software draws the peaks and adds them up. If the main peak is 99% of that sum, the certificate says 99%.

So the figure is a proportion of the detected sample, not of everything present. The peaks alongside the main one are usually peptide-related impurities from synthesis or storage: sequences missing an amino acid, sequences with an extra one, oxidised or deamidated forms, and diastereomers. D'Hondt and colleagues (2014) catalogued these as the impurities most commonly seen in peptide medicines, and they are what a good HPLC purity run is designed to resolve (see where impurities come from).

What the method cannot see is simply not in the denominator. Water, counter-ions such as acetate or trifluoroacetate, residual solvents and endotoxin are not chromatographic peaks in a standard peptide purity run at peptide wavelengths, so they can be entirely absent from a 99% figure.

Why 99% is not 99% by mass

Because those non-peptide components carry no signal, a purity percentage is not a mass fraction. A peptide can be 99% pure by HPLC while a meaningful share of its weight is water and counter-ion. Wang and colleagues (2022) showed the size of the gap when they assigned a reference value to a synthetic oxytocin by mass balance: after measuring water by Karl Fischer titration, counter-ions by ion chromatography and residual solvents by gas chromatography, the peptide content came out at 796.5 mg per gram, even though the material was of ordinary pharmaceutical quality by chromatography.

There is a second, smaller effect inside the chromatogram itself. Area percentages assume every component absorbs UV light equally per unit of mass. They do not. ICH Q2(R2), the harmonised guideline on validating analytical procedures, addresses this directly: where an impurity responds differently from the main compound, a relative response factor is calculated, and if it falls outside 0.8 to 1.2 a correction factor is applied. Many routine certificates report uncorrected area normalisation, which is acceptable for a purity test but is one more reason the number is a proportion of signal rather than of substance.

A percentage without a method is not a result

Purity is method-dependent. The same sample can produce different figures under different conditions: a different column, gradient, mobile phase, wavelength or run time. An impurity that co-elutes with the main peak in one method may separate in another; a hydrophobic aggregate may elute after the gradient ends and never be integrated. A purity figure is only comparable when the method it came from is stated alongside it.

This is why ICH Q2(R2) treats a purity procedure as something that has to be validated for its purpose, with specificity, precision and range demonstrated for the actual conditions used. The useful form of a purity result therefore states the method (see what HPLC and mass spectrometry each establish), the specification the batch was tested against, and the measured figure.

Purity is one question among several

Behind any peptide sit separate questions: what is it, how pure is it, and is it sterile. Purity answers one. A high purity figure says nothing about whether the pen contains the labelled compound, because HPLC compares peaks by retention time rather than confirming structure, and nothing about sterility, which matters most for a product presented for injection.

Every batch we sell is tested by an independent laboratory for exactly those three things: identity by mass spectrometry, purity by HPLC, and sterility. The certificate for each batch is published at our COA index as a plain PDF, and the QR code on the pen opens the same document. Take the MOTS-C 32 pen as an example: the certificate you reach from its QR code is the exact-lot report, not a generic sample.

Pharmacopoeial monographs for licensed peptide medicines go further. A full release specification typically adds a content or assay figure against a calibrated reference standard, water content, counter-ion content, residual solvents and an endotoxin limit, which is how the mass-balance study above was assembled. Those are additional tests within a regulatory dossier, and their absence from a certificate does not make the tests that are on it less real. It does mean the purity figure should be read for what it is: the chromatographic result for that batch under the stated method.

Reading the certificate alongside the parcel

A certificate documents the batch at the time of laboratory testing. It cannot say anything about what happened afterwards, which is why we ship in insulated packaging with a cooling element and a temperature indicator, holding the pens at 2–8 °C from our Czech dispatch point to your refrigerator. The indicator answers the transit question; the certificate answers the release question. Check both when the parcel arrives: look at the indicator (batch and lot matching covers the certificate side). If it shows a breach, we replace the pen free of charge after a photo or short video, whatever the certificate says, because the certificate cannot retrospectively prove stability under conditions it never measured.