How HPLC purity is measured for peptides

What does HPLC measure in a peptide sample?

HPLC measures how much ultraviolet light, the invisible light just past violet, each separated component of a dissolved sample absorbs as it leaves the instrument. The instrument pushes the mixture through a column, a packed tube, so that each component leaves at its own time and draws its own peak on the chromatogram. McCarthy et al. describe reverse-phase HPLC, in which a water-repelling column lining holds back longest the molecules that most avoid water, with peptides watched at 220 nanometres (nm, a billionth of a metre) of ultraviolet light (McCarthy et al., Pharmaceutical Research, 2023). The detector records absorbed light only. It does not read a name, a weight or a sequence, the order of amino acids.

A component's retention time is the delay between loading the sample and seeing its signal. A peak's position tells when something left, and a peak's height and width tell how much light it absorbed. Verbeke et al. read their peptides at 210 and 215 nm in two setups, and the supplier method they checked for one peptide used 220 nm (Verbeke et al., Journal of Pharmaceutical Analysis, 2015).

How is HPLC purity calculated from a chromatogram?

HPLC purity is calculated by area normalisation: the area under the main peak, the space between its curve and the flat line beneath it, is divided by the total area of all detected peaks and written as a percentage. Verbeke et al. quantified peptide purity as the area of the most abundant peak as a percentage of total peak area (Verbeke et al., 2015). McCarthy et al. report the other side of the same sum, impurities as a percentage of total detected area (McCarthy et al., 2023). ICH, the body behind shared drug-testing guidelines, asks that the approach be described, for example area percent (ICH Q2(R2), 2023).

The main peak and the other peaks share one total of 100%, so a 99% main peak leaves 1% of the area for everything else the detector saw. Anything the chromatogram does not record sits outside the sum, which is why McCarthy et al. apply a second step that removes water, counter-ions (oppositely charged small molecules that cling to the peptide), leftover solvents (liquids used in making it) and inorganic residue (mineral ash left after burning), each measured by weight (McCarthy et al., 2023).

What does 99% purity mean?

A 99% HPLC purity figure means that, in one run on one method, the main peak held about 99% of the total detected peak area and all other peaks together held about 1%. The figure describes a share of signal, not a share of weight, and a different column, solvent mix (the liquid that carries the sample) or wavelength (the exact kind of light the detector reads, measured in nm) can give a different figure for the same material. Verbeke et al. show the gap: a supplier method put one peptide at 95.0% purity or more, while the researchers measured 80.6% on their own system (Verbeke et al., 2015).

The number 99 is a result, not a grade. A report without the method gives a reader no way to compare it with another lab's figure. ICH Q2(R2) asks that a test method be shown to suit its purpose, including specificity, the extent to which other substances interfere with measuring the target (ICH Q2(R2), 2023).

Why can two labs report different purity for the same peptide?

Two labs can report different HPLC purity for one peptide because a detector does not respond equally to every molecule, and because a method may not separate every impurity from the main peak. Kuipers and Gruppen measured how strongly peptide parts absorb light at 214 nm and found that tryptophan, an amino acid, absorbs about 30 times more than the peptide bond, the link joining amino acids (Kuipers and Gruppen, Journal of Agricultural and Food Chemistry, 2007). ICH Q2(R2) tells analysts to calculate a relative response factor, the ratio between how strongly an impurity and the main substance register on the detector, when responses differ, and to apply a correction outside 0.8 to 1.2 (ICH Q2(R2), 2023).

Co-elution is the second cause: two molecules leave the column together and draw one peak. The EMA names the risk of co-eluting impurities in its December 2025 synthetic-peptide guideline, and says that when they appear as one peak the 1.0% qualification threshold, the level above which an impurity needs safety data, applies unless justified (EMA/CHMP/CVMP/QWP/367182/2025). Verbeke et al. judged that the supplier method for the 80.6% peptide lacked selectivity, the power to separate the main peak from its neighbours (Verbeke et al., 2015).

Does HPLC purity show identity, amount or safety?

HPLC purity does not show identity, amount or safety, because each needs a separate test. Identity asks whether the main peak is the named peptide, and the EMA synthetic-peptide guideline recommends at least two orthogonal methods, meaning tests built on different principles, with mass (molecular weight) among the accepted ones (EMA/CHMP/CVMP/QWP/367182/2025). The same guideline lists assay, or content (how much of the named peptide is present), as its own specification item, one of the fixed tests a batch must pass, apart from purity. It also lists bacterial endotoxins (poisons released by bacteria) and microbiological purity (freedom from living microbes) as further separate tests, so a high area percentage leaves those questions open.

A purity reading can even count the wrong molecule. In the Verbeke study, the main peak of one peptide was not the requested sequence: its mass differed by 2.1 daltons (Da, a unit of molecular mass) from the target (Verbeke et al., 2015). A sequence implies a mass, which is why the retatrutide molecular-formula record shows how a mass can be checked.

Evidence limits

What has research not shown about HPLC purity?

The studies are narrow too. Verbeke et al. tested 98 synthetic quorum-sensing peptides, signalling molecules that bacteria use, and found 43 met the requested 95.0% purity (Verbeke et al., 2015). The McCarthy paper describes reference standards, the carefully tested yardsticks that other tests are weighed against, not commercial lots (batches made for sale), across six peptides (McCarthy et al., 2023). Neither tests any 2026 catalogue. A purity figure also says nothing about permission to supply or use a material, and research-use-only material is not a medicine and is not for human or animal use. For paperwork more widely, see what research grade actually means.