HPLC vs Mass Spec: Two Different Questions

A peptide sample can be 99% pure and still be the wrong molecule. Those two statements do not contradict each other, because they answer different questions, and the instruments that answer them are not interchangeable. High-performance liquid chromatography, or HPLC, reports a proportion: of everything the detector registered, what share was a single component. Mass spectrometry reports something closer to an identity: what that component weighs, and therefore which molecule it most likely is. A seller can hand you an excellent HPLC figure for a batch of something you never ordered, and nothing in that figure will say so. What follows is where the line between the two techniques sits, and what neither of them settles. It is educational context, not medical advice, and it recommends nothing.

What the purity percentage is a percentage of

In the reverse-phase mode peptide laboratories run, HPLC separates a sample by how strongly each component sticks to a hydrophobic column packing while a changing solvent gradient washes it back off. Components that leave at different times register as separate peaks. Detection is almost always by ultraviolet absorbance in the far ultraviolet, around 220 nm, because the peptide bond itself absorbs strongly there. Purity is then calculated by area normalisation: the area of the main peak divided by the total area of every peak, expressed as a percentage. Every limitation follows from that definition. The number is a ratio between things the detector registered. It is silent about anything the detector never registered, silent about two components hiding beneath a single peak, and entirely silent about what the main peak is. A chromatogram of a well-made 43-residue protein and a chromatogram of a well-made seven-residue fragment can both be one sharp peak at 99% area.^[6]

What a mass spectrometer answers instead

A mass spectrometer has no interest in how long something took to come off a column. It ionises the molecule and measures its mass-to-charge ratio, and a peptide's mass is a direct consequence of its amino acid sequence. Swap one residue for another, lose one, gain an extra one, oxidise a methionine, and the mass moves by a predictable amount. That is why identity work belongs to mass spectrometry rather than to chromatography. Tandem mass spectrometry, written MS/MS, goes further again: it breaks the peptide along its backbone and measures the fragment masses, which lets an analyst reconstruct the sequence rather than merely confirm a total. In the published work behind one pharmacopoeial peptide reference standard, high-resolution mass spectrometry confirmed the molecular mass of leuprolide, and MS/MS then gave complete coverage of its amino acid sequence. Those are two separate confirmations, and the second is much the stronger of them.^[7]

The TB-500 case: a clean peak of the wrong molecule

The clearest worked example in this market is TB-500, which is sold alongside citations to research on thymosin beta-4, a 43-residue protein human cells make for themselves. In 2012 a doping-control laboratory at Ghent University analysed a commercial TB-500 formulation by high-performance liquid chromatography coupled to high-resolution mass spectrometry and identified what was in it: the N-terminal acetylated 17-23 fragment of human thymosin beta-4, seven residues, Ac-LKKTETQ. The same team then synthesised that fragment independently to confirm the assignment. Seven residues is not 43 residues. A purity assay would have been perfectly content with either, because a carefully synthesised seven-residue peptide is a clean single peak. Only the mass told the analysts which molecule was actually on the bench. Our data sheet at TB-500 covers what that identity difference does to the research people cite for it.^[1]

Three ways a tidy chromatogram overstates purity

The first is co-elution. The impurities peptide synthesis produces are usually close relatives of the target: a chain missing one residue, a chain carrying one extra, an oxidised or deamidated version. Close relatives behave similarly on a column, so they arrive at nearly the same moment and their peaks overlap. Analysts at the FDA's own Division of Pharmaceutical Analysis reported precisely this, describing a study in their laboratory that found both an amino acid deletion and an amino acid insertion in a synthetic peptide drug product which the submitted HPLC-UV methods could not separate, and noting more bluntly that many of the quality control methods submitted for peptide drugs do not appear to be adequate for distinguishing and quantifying such impurities. The second problem is ultraviolet blindness. Area normalisation assumes every component absorbs, and some do not. Authors of a 2020 mass balance study of synthetic glucagon attributed the gap between their figure and the chromatographic one to exactly that: some impurities do not absorb ultraviolet radiation, while others have absorption features overlapping those of the main constituent. The third is the point of this whole page. Area normalisation holds no opinion about identity. It compares peaks against each other, and it would report the same percentage for the right peptide and for a different peptide made just as well.^{[2, 3]}

Mass spectrometry has its own blind spot

It would be dishonest to present mass spectrometry as the answer to everything, because it fails on one specific and important class of impurity: isomers. Two molecules assembled from the same atoms in the same numbers weigh the same, whatever order those atoms sit in. The textbook case in peptide chemistry is aspartic acid isomerising to isoaspartic acid, a rearrangement of the backbone connection that leaves the mass untouched. A 2022 paper on the problem calls it a common isobaric impurity that can be very difficult to identify without first synthesising isoAsp standards for chromatographic comparison, and notes that ordinary collision-induced fragmentation yields fragment ions which themselves remain isobaric and are therefore ambiguous. Telling them apart takes specialised approaches such as ion mobility separation. So the honest summary is not that mass spectrometry wins. It is that each instrument is blind in a different direction, which is exactly why careful analytical work runs both.^[4]

What regulators ask for when the stakes are high

The contrast with regulated medicine is instructive. The FDA's 2021 guidance on abbreviated applications for certain synthetic peptide products encourages applicants to bring orthogonal analytical methods to bear on primary sequence and physicochemical properties, among other attributes. Orthogonal is the operative word: methods that fail in different directions, so that one covers what another misses. The same document sets the expectation that a proposed generic carries no new specified peptide-related impurity above 0.5 percent of the drug substance, and that impurities between 0.10 percent and 0.5 percent be identified, which the document defines as characterising the structure of the impurity. Structure, not peak area. None of this reaches the research-peptide market, which sells material that is not FDA-approved and never enters that process at all. The comparison is still worth having, because it shows what the parties with the most to lose treat as sufficient, and a lone area percentage is not on that list.^[5]

Why control laboratories reach for the mass first

There is a practical tell in how enforcement laboratories work. When Belgian medicines and customs authorities send suspect injectable peptide preparations to an official medicines control laboratory, the screening method those analysts published is liquid chromatography with tandem mass spectrometry, able to detect 25 different peptides and framed around the minimum of five identification points recommended for sports drug testing. The chromatography is there to separate. The mass spectrometry is there to decide what the separated thing is, and the identification-point framework exists because deciding identity is the difficult half. The set of peptides that method covers was assembled from substances already found in illegal and counterfeit products seized across European countries. When the question is what is inside this container, nobody answers it with an area percentage.^[8]

Reading a certificate with both questions in mind

Once the two questions come apart, a certificate reads differently. Everything beneath the purity heading answers how much of one thing. Everything beneath the identity heading answers what thing. A sheet that shows a strong chromatographic figure and nothing else has answered one of the two, and it is not the one that tells you whether you received what you ordered. The field-by-field walkthrough lives at how to read a peptide COA, and the separate question of who ran the test in the first place is covered at what third-party tested actually means. Neither is repeated here. What this page adds is the reason both techniques belong on the same sheet: they are not redundant, and neither is a stronger version of the other.

The one-line version

HPLC answers how much of one thing is in the sample, mass spectrometry answers what that thing is, and a certificate reporting only the first has told you roughly half of what it appears to have told you. That is a limit on the quality of your evidence rather than a statement about the compound, which is whatever it is regardless of what anyone measured. Purity percentages do a great deal of selling in this category. Knowing which question the number answers costs nothing and changes how you read every sheet you are shown afterwards. There is a third question the same certificate rarely answers, which is how much of the weight in the container is peptide at all rather than counterion and water; that one is covered at purity vs net peptide content. For the wider context on what this material is and how it is sold, what are research peptides is the place to start.

References & sources

1. Esposito S, Deventer K, Goeman J, Van der Eycken J, Van Eenoo P. Synthesis and characterization of the N-terminal acetylated 17-23 fragment of thymosin beta 4 identified in TB-500, a product suspected to possess doping potential. Drug Testing and Analysis (2012). Ghent University doping-control laboratory; HPLC with high-resolution mass spectrometry identified Ac-LKKTETQ in a commercial TB-500 formulation, which the authors then synthesised independently.
2. Zeng K, Geerlof-Vidavisky I, Gucinski A, Jiang X, Boyne MT 2nd. Liquid Chromatography-High Resolution Mass Spectrometry for Peptide Drug Quality Control. The AAPS Journal 17(3):643-651 (2015). Division of Pharmaceutical Analysis, CDER. Reports that many HPLC-UV quality control methods submitted for peptide drugs do not appear adequate for distinguishing and quantifying process impurities, and that a study in the authors' laboratory found an amino acid deletion and an amino acid insertion that the submitted HPLC-UV methods could not separate.
3. Wang X, Zhang F, Li H, Xiao P, Su F, Xu B, Sun W, Song D. Purity determination of synthetic glucagon using a mass balance approach. Scientific Reports (2020). States that some impurities do not absorb ultraviolet radiation while others have absorption features overlapping the main constituent, and that peptide impurities with sequences close to the main component have retention times close to it.
4. Gibson K, Cooper-Shepherd DA, Pallister E, Inman SE, Jackson SE, Lindo V. Toward Rapid Aspartic Acid Isomer Localization in Therapeutic Peptides Using Cyclic Ion Mobility Mass Spectrometry. Journal of the American Society for Mass Spectrometry (2022). Describes aspartic acid isomerisation to isoaspartic acid as a common isobaric impurity that is very difficult to identify without synthesised isoAsp standards, and notes that collision-induced dissociation yields fragment ions that remain isobaric and ambiguous.
5. U.S. FDA. ANDAs for Certain Highly Purified Synthetic Peptide Drug Products That Refer to Listed Drugs of rDNA Origin. Guidance for Industry, May 2021. Encourages orthogonal analytical methods for primary sequence and physicochemical properties; sets an expectation of no new specified peptide-related impurity above 0.5 percent of the drug substance, with impurities of 0.10 percent to 0.5 percent identified, meaning their structure characterised.
6. Mant CT, Chen Y, Yan Z, Popa TV, Kovacs JM, Mills JB, Tripet BP, Hodges RS. HPLC Analysis and Purification of Peptides. Methods in Molecular Biology, vol. 386, pp. 3-55 (2007). States that peptide bonds absorb strongly in the far ultraviolet, around 220 nm, which is the basis for the usual 210 to 220 nm detection range.
7. McCarthy D, Han Y, Carrick K, Schmidt D, Workman W, Matejtschuk P, Duru C, Atouf F. Reference Standards to Support Quality of Synthetic Peptide Therapeutics. Pharmaceutical Research 40(6):1317-1328 (2023). For leuprolide, high-resolution mass spectrometry confirmed the molecular mass and MS/MS gave complete coverage of the amino acid sequence.
8. Vanhee C, Janvier S, Desmedt B, Moens G, Deconinck E, De Beer JO, Courselle P. Analysis of illegal peptide biopharmaceuticals frequently encountered by controlling agencies. Talanta 142:1-10 (2015). An LC-MS/MS screening method run at a Belgian official medicines control laboratory at the request of the national medicines agency and customs, selectively detecting 25 peptides and incorporating the minimum of five identification points recommended for sports drug testing; the peptide set was assembled from substances already found in illegal and counterfeit products seized by European countries.