Net Peptide Content vs HPLC Purity: The Two Numbers on a COA That Are Not the Same Number

They are ratios with different denominators, and that is the whole answer. HPLC purity asks: of the peptide-related material the detector saw, what fraction was the target sequence? Net peptide content asks: of the total mass weighed out of the vial, what fraction is peptide at all? A single lot of lyophilised material can honestly report 98.5% chromatographic purity and 78% net peptide content on the same certificate, because the first number never counted the counterion, the water, the residual solvent or the inorganic residue that make up the rest of the powder. USP General Chapter 〈1503〉, which sets out the quality attributes expected of synthetic peptide drug substances, treats peptide content and assay, counter-ion content, water content and related-substance impurities as separate attributes with separate tests for exactly this reason ^[1].

The divergence is structural rather than a defect in either method. Chromatographic purity is computed by area normalisation: the area of the main peak divided by the total integrated area of all peaks. Every term in that fraction is a peak. Anything in the powder that produces no peak — because it does not absorb at the detection wavelength, because it elutes unretained with the injection front, or because it never comes off the column — sits in neither numerator nor denominator, and is simply invisible to the calculation. Net peptide content is a mass fraction, and mass fractions have no blind spot: whatever is in the vial and is not peptide counts against it.

What HPLC purity actually measures

On peptide material the conventional purity method is reversed-phase HPLC with ultraviolet detection near 214 nm, where the peptide bond itself absorbs, or near 220 nm on some methods. Detecting at the amide backbone rather than at 280 nm is deliberate: 280 nm responds mainly to tryptophan, tyrosine and cystine, so a sequence containing none of them would be close to invisible. The result is reported as area percent, and it is a sound measurement of one specific thing — the proportion of the eluting, UV-absorbing, peptide-related population that is the target sequence rather than a deletion sequence, a truncation, an oxidation product, an epimer or a protecting-group adduct.

Area percent carries an assumption worth stating out loud rather than inheriting: that every species in the chromatogram responds equally per unit mass. At 214 nm that is approximately true for closely related peptide impurities, which is why the method holds up at all — a deletion sequence missing one residue absorbs much like the parent. It is considerably less true across compositional differences. Measured against the peptide bond, the molar absorptivity of tryptophan at 214 nm is roughly thirty times higher; phenylalanine, tyrosine and histidine roughly six times; proline about three times ^[6]. An impurity that has lost a tryptophan is therefore under-represented in area percent, and one that carries an extra aromatic residue is over-represented. The bias is often small. It is not always small, and it is never zero.

Two further limits belong in the same breath. Detection at 214 nm is not selective for peptides — residual reagents, scavengers and solvent-related species absorb there too, and unless the procedure has been shown to separate them they can be integrated as though they were peptide impurities ^[6]. And co-elution is undetectable by definition: two species under one peak report as one peak. ICH Q6A requires that the procedure used to assay a drug substance be specific and stability-indicating, and treats a single chromatographic procedure as serving both assay and impurity quantitation only where it has been shown to do both ^[9].

What net peptide content measures

Net peptide content — printed variously as peptide content, net content or assay — is the mass of peptide expressed as a percentage of the mass of material weighed. It is the figure that determines how much peptide a given gross weight actually contains. USP 〈1503〉 lists content and assay among the core quality attributes for synthetic peptide drug substances, alongside counter-ion content, water content, residual solvents and inorganic impurities, each with its own determination ^[1].

There are two routes to it. The first is mass balance: measure everything that is not the target peptide and subtract. In the value-assignment strategy USP published for peptide reference standards, purity is calculated as the chromatographic result combined with the non-peptide mass remaining after counter-ion, residual solvent and inorganic residue are deducted, then corrected for water measured separately in each participating laboratory ^[5]. Their worked illustration reads better as arithmetic than as theory: 0.92% chromatographic impurities, 5.58% w/w acetic acid, 0.003% w/w trifluoroacetic acid, 0.003% w/w residual solvents and no detectable residue on ignition combine to 0.93 mg of peptide in each milligram of material, which a water correction of 1.5% or 2.0% — depending on which laboratory ran it — moves to 0.92 or 0.91 ^[5]. A chromatographic purity above 99% and a content near 92% are describing the same lot without contradicting each other.

The second route is quantitative amino acid analysis, which measures peptide mass directly rather than by difference. The sample is hydrolysed — conventionally in 6 M hydrochloric acid at about 110°C for 24 hours — and the released amino acids are separated and quantified against an internal standard, yielding both composition and total amino acid mass. USP 〈1052〉 is the harmonised chapter describing the methodology, including the requirement that interfering buffer components such as salts, urea and detergents be removed before analysis ^[2].

Amino acid analysis has documented weaknesses that certificates rarely mention. Tryptophan is destroyed by acid hydrolysis; cysteine is largely lost unless derivatised beforehand; serine and threonine degrade progressively, so their values are commonly extrapolated back to zero hydrolysis time; and bonds between adjacent β-branched residues such as isoleucine and valine hydrolyse slowly, leaving those residues under-recovered at 24 hours ^[2]. There is also a conceptual limit that matters more than any of them: amino acid analysis measures amino acids, so a deletion sequence or a truncated fragment contributes to the total exactly as the target does. Content by amino acid analysis and purity by chromatography are therefore complementary rather than redundant. Content says how much peptide is in the powder; purity says what fraction of that peptide is the right one. Neither substitutes for the other.

Why the counterion is mass, not a footnote

The largest single non-peptide component of a typical lyophilised powder is usually the counterion. Reversed-phase purification is conventionally run with trifluoroacetic acid as the ion-pairing additive, and the peptide leaves that process as a trifluoroacetate salt, the anion associating with protonated basic sites: arginine and lysine side chains, histidine, and the free N-terminus. The mass consequence is arithmetic. Trifluoroacetate has a formula mass near 113 and acetate near 59, so a sequence carrying several basic residues carries several multiples of that mass in counterion before water is considered at all. Two peptides of identical chromatographic purity can differ materially in net content purely because one has more basic residues. USP publishes separate general chapters for the two determinations — 〈503〉 for acetic acid ^[3] and 〈503.1〉 for trifluoroacetic acid ^[4] — because they are different analytes requiring different methods, and because counter-ion content is a specification in its own right ^[1].

Salt form is also changeable, and changing it is not free. Roux and colleagues compared three routes for removing or exchanging trifluoroacetate on a dicationic peptide — repeat reversed-phase chromatography with a different acid, ion-exchange resin, and a deprotonation followed by reprotonation of the amino groups — and reported that the approaches differ substantially in completeness and in recovered material, with residual trifluoroacetate persisting in more than one of them ^[8]. The reason anyone goes to that trouble is that trifluoroacetate is not an inert spectator on the bench. In cultured fetal rat osteoblasts and articular chondrocytes, and in neonatal mouse calvarial organ culture, trifluoroacetate reduced cell number and thymidine incorporation at low nanomolar concentrations ^[7]. That evidence is in vitro and ex vivo in rodent tissue, not a human study, and it should not be stretched past what it shows. What it does establish is that the counterion can act as a variable in a cell-based assay rather than as packaging. A certificate that does not state the salt form leaves an unrecorded reagent in the experiment.

Reading the two numbers together

Put the document in front of you and read the wording, not the size of the font. Purity and content are two entries among many, and the rest of the document repays a field-by-field read of the certificate. The label beside each number carries most of the meaning, and the method beside each label carries the rest.

• The word beside the number. Purity, chromatographic purity or HPLC area percent is a ratio among peaks. Peptide content, net peptide content or assay is a mass fraction. Where only one of the two appears, only one of the two questions has been answered ^[1].
• The method beside each number. A purity figure should name the technique, the detection wavelength and the separation conditions; a content figure should say whether it came from mass balance or from amino acid analysis, because those two carry different blind spots ^{[2] [5]}.
• The salt form, stated explicitly as trifluoroacetate, acetate or hydrochloride, with a measured counter-ion percentage rather than an unquantified claim ^{[3] [4]}.
• Water content with its method named — Karl Fischer titration is the usual one — since residual water is normally the second-largest non-peptide mass in a lyophilised powder ^[1].
• Residual solvents and inorganic residue, or an explicit statement that they were tested and not detected ^[5].
• Internal consistency. Content, counterion, water, solvents and residue should account for approximately the whole mass. If they sum well short of 100% with nothing named to fill the gap, something was not measured ^[5].

It is worth being plain about the state of practice, because doing otherwise would repeat the failure this article is about. Most certificates circulating for research-grade peptide material report chromatographic purity and nothing else. Net peptide content appears far less often, and when it does the method behind it is frequently unstated. The mass-balance approach USP developed was built for reference standards produced under controlled conditions and characterised across several collaborating laboratories — and even there, the published work documents inter-laboratory variability in the water correction and in the assay results being combined ^[5]. A content figure with no method behind it is not a strong number, and the honest entry in a receiving log for an absent or unattributed one is "not reported" rather than an assumed value. Any later calculation depending on peptide mass inherits that gap whether or not it is written down.

What neither number tells you

Neither purity nor content is an identity test. A chromatogram establishes that one dominant species eluted at a particular retention time. A content assay establishes that a certain mass of amino acids was present. Neither demonstrates that the sequence is the sequence printed on the label. Identity is established orthogonally — most commonly by mass spectrometry, comparing an observed mass against the theoretical mass for the stated sequence, and where necessary by peptide mapping. ICH Q6A is explicit that identity, assay and impurity tests are distinct items in a specification, each with its own analytical procedure and its own acceptance criterion ^[9]. A certificate reporting a purity percentage and nothing else has not established identity, however high the percentage is.

Both numbers also share a limit no analytical rigour removes: each describes a sample drawn from a lot on a date, by a stated method, at a named laboratory. Extending either to the container in front of you is an inference resting on the lot string matching the label and the dates being coherent — which is why the document-matching checks come first and the percentages come second. A precise number attached to unverifiable material is not more informative than no number at all. It is less, because it invites confidence nothing in the document supports.