Peptide Reference Standards and Why Laboratories Disagree on Peptide Purity

Peptide reference standards are the anchor that makes a purity number mean something. A percentage on a certificate is not a property of the vial, it is the output of one method calibrated against one standard, and international comparison studies show that expert laboratories measuring the same synthetic peptide can arrive at materially different answers. Without a stated method and a defined reference standard, a purity figure is an assertion rather than a measurement.

Most discussion of research peptide quality stops at a single number. A supplier publishes 98.7 percent, a buyer reads 98.7 percent, and the transaction proceeds. What that exchange omits is the entire measurement chain that produced the figure: which method was run, what the instrument was calibrated against, how the calibrator itself was valued, and how much of the remaining mass was never examined at all. Metrology, the science of measurement, treats those questions as the substance of the result. Analytical chemistry applied to synthetic peptides makes them unavoidable, because a synthetic peptide is never a single substance.

What a Peptide Reference Standard Actually Is

A reference standard is a characterised material whose peptide content has been assigned by a defined procedure, against which a routine assay is calibrated. The hierarchy runs from primary calibrators, whose values are assigned by methods that do not depend on a like calibrator, down to secondary and in-house working standards that inherit their value by comparison. Quantitative nuclear magnetic resonance against a certified internal standard and mass balance calculation both function as primary approaches. Everything downstream borrows credibility from them.

This hierarchy is why the phrase “calibrated against a reference standard” carries so little information on its own. If the working standard is simply a retained lot of the supplier’s own bulk material, the assay measures consistency between batches rather than absolute content. That is a useful quality control signal and a poor answer to the question of how much peptide is present. The distinction is invisible on a certificate that reports only a percentage.

Purity and Content Are Different Questions

Chromatographic purity, typically reported as area percent from reversed phase separation with detection at 214 nm where the amide bond absorbs, answers a narrow question: of the ultraviolet absorbing material that eluted from the column, what fraction was the target peptide. Content, sometimes called assay or net peptide content, answers a different one: of the total mass in the vial, how many milligrams are peptide.

The gap between those two numbers is occupied by material that area percent cannot see. Counterion from the final purification step, residual water bound to the lyophilised cake, residual organic solvent, and inorganic salts all contribute mass while contributing little or nothing to the chromatogram at 214 nm. A material can be 99 percent pure by area and still be substantially less than 99 percent peptide by mass, and both statements can be simultaneously correct. Our discussion of residual trifluoroacetic acid and counterion content covers the largest single contributor to that gap.

Key Research Findings

• In the CCQM-P55.2.2018 pilot study on peptide purity coordinated by the Bureau International des Poids et Mesures, the reference value for the mass fraction of the target hexapeptide in the study material was 613 mg/g with a standard uncertainty of 10 mg/g, meaning the material was approximately 61 percent peptide by mass (Josephs et al., BIPM Final Report, April 2022).
• In the same study material, trifluoroacetic acid accounted for 286.7 mg/g with a standard uncertainty of 2.3 mg/g, water for 47.5 mg/g with an uncertainty of 4.1 mg/g, and peptide related impurities for 53.0 mg/g with an uncertainty of 8.6 mg/g.
• All five participating national metrology institutes underestimated total peptide related impurities relative to the reference value of 53.0 mg/g. Degrees of equivalence were -30.0 mg/g for NIM, -34.1 mg/g for BIPM by impurity corrected amino acid analysis, -34.1 mg/g for BIPM by impurity corrected quantitative NMR, -23.0 mg/g for TUBITAK UME, and -28.0 mg/g for KRISS, against expanded uncertainties of 17.3 to 17.5 mg/g at approximately 95 percent confidence.
• Every one of those deviations exceeded its own expanded uncertainty, indicating systematic rather than random disagreement. The report attributes the pattern to a single shared cause: all participants failed to identify and quantify the major related impurity, a depsipeptide of the target sequence.
• In a United States Pharmacopeia multi-laboratory collaborative study comparing HPLC assay, quantitative NMR and amino acid analysis for quantitation of the nonapeptide oxytocin, the HPLC assay using the same peptide bulk material as the standard showed the lowest inter-laboratory variability (Li et al., Journal of Pharmaceutical and Biomedical Analysis, 2019, volume 166, pages 105 to 112).
• That reported coefficient of variation was calculated without including the uncertainty associated with the mass balance purity assignment of the standard itself, so the apparent advantage of the HPLC approach depends on an input whose own uncertainty was excluded from the comparison.

What the CCQM Comparisons Revealed

The Consultative Committee for Amount of Substance runs formal comparison exercises in which national metrology institutes independently measure the same distributed material. These are the most rigorous available tests of whether peptide purity measurement is reproducible across laboratories, because the participants are the institutions that define measurement for their countries and because the sample is identical.

The CCQM-P55.2.2018 pilot study distributed a synthetic hexapeptide derived from HbA0, coordinated by Ralf Josephs at the BIPM in Sèvres with coordination laboratories at the Health Sciences Authority in Singapore and the National Institute of Metrology in Beijing, and contributions from KRISS in Korea and TUBITAK UME in Turkey. The headline figure is worth sitting with. The reference value for the peptide itself was 613 mg/g. Nearly 39 percent of the distributed material was something other than the target peptide, and the largest component of that remainder was counterion at 286.7 mg/g.

When Expert Laboratories Disagree Systematically

The more instructive result concerns the impurity measurement rather than the content measurement. Five independent results were submitted for total peptide related impurities. All five fell below the reference value of 53.0 mg/g, and all five missed by more than their own stated expanded uncertainty. Random error does not produce that pattern. A shared blind spot does.

The final report identifies the cause directly: every participant omitted the depsipeptide impurity, an ester linked variant arising from the synthesis, because their analytical approaches did not resolve or detect it. The impurity was eventually identified by one laboratory in the parallel key comparison and confirmed by two others. This is the practical lesson that no purity percentage can carry on its own. An impurity that a method cannot see does not appear in the result, and its absence is indistinguishable from a clean sample unless someone goes looking with a different technique. Method validation under ICH Q2 criteria establishes that a method performs reliably within one laboratory, which is necessary but does not by itself guarantee agreement between laboratories.

Which Method Assigns Content Best

The United States Pharmacopeia addressed the method selection question directly through its Biologics Department, in a multi-laboratory collaborative study comparing three approaches to quantifying oxytocin. Li and colleagues reported the comparison in the Journal of Pharmaceutical and Biomedical Analysis in 2019, with authorship spanning USP, the National Research Council of Canada, Health Canada, the Swedish Medical Products Agency, and the National Institute for Biological Standards and Control among others.

HPLC assay run against the same peptide bulk material as the standard produced the lowest inter-laboratory variability. That result is less surprising than it first appears, and the authors are careful about why. When every laboratory calibrates against the same physical material, a systematic error in the assigned value of that material propagates identically to every laboratory and therefore cancels out of the between-laboratory comparison. Agreement improves without accuracy necessarily improving. The reported coefficient of variation explicitly excluded the uncertainty of the mass balance purity assignment of the standard.

Quantitative NMR occupies a different position in the hierarchy. Because it measures the target directly against a certified internal standard rather than against a like material, it does not inherit an upstream value assignment, and the authors concluded it deserves further exploration as a primary method for peptide reference standard value assignment. The tradeoff is that it requires a spectrometer, a suitable certified internal standard, and non trivial method development. Our overview of quantitative NMR for absolute peptide content covers the technique in more depth.

What Traceability Means for a Research Peptide Certificate

Metrological traceability is the property of a result being relatable to a stated reference through a documented, unbroken chain of calibrations, each contributing to measurement uncertainty. Applied to a research peptide certificate, it reduces to a small set of questions that a well constructed document answers and a weak one does not.

The first is which method produced the number, since purity by area percent and content by mass are not interchangeable and a certificate reporting only one has answered only one question. The second is what the instrument was calibrated against and how that calibrator was valued, since a working standard qualified against a supplier’s own retained bulk provides batch consistency rather than absolute content. The third is what the total mass balance looks like, because counterion, water and residual solvent are frequently the largest non peptide components and are invisible to chromatographic purity. The fourth is whether the certificate is batch specific and traceable to the vial in hand, since a certificate for a different lot describes a different material.

Maple Research Labs publishes third party analytical documentation through Janoshik Analytical and reports the method alongside the result rather than a bare percentage. Product listings such as BPC-157 link the supporting analytical report for the specific batch so that the number can be traced to the material it describes.

Limitations and Open Questions

The comparison studies discussed here examined well characterised model peptides. Oxytocin is a nonapeptide with a defined disulfide bridge and decades of pharmacopeial history. The HbA0 hexapeptide was selected for a metrology exercise. Longer sequences, sequences prone to aspartimide formation or racemisation, and materials with complex post synthetic modification present harder analytical problems, and the interlaboratory spread observed on model compounds should be read as a lower bound rather than a typical case.

A second limitation is that both studies enrolled national metrology institutes and pharmacopeial laboratories. These are the best resourced analytical laboratories available, working to documented protocols on a distributed sample with the explicit knowledge that their results would be compared. The depsipeptide finding is more striking for that reason, not less. It also means the observed disagreement is not a reasonable estimate of the spread across commercial testing laboratories operating without that scrutiny.

Finally, the mass balance approach that underpins primary value assignment depends on quantifying every non peptide component, which in practice means water by Karl Fischer, counterion by ion chromatography or nuclear magnetic resonance, residual solvent by gas chromatography, and inorganic residue by elemental analysis. Each contributes its own uncertainty, and an unmeasured component is silently assigned to the peptide. The 613 mg/g figure from CCQM-P55.2.2018 is the product of that full accounting, which is precisely why it sits so far below what a chromatographic purity figure alone would have suggested.

For research purposes only. Not for human consumption. Not for diagnostic or therapeutic use.