Peptide Cross-Contamination: The Impurity a COA Purity Number Cannot See

Peptide cross-contamination is the presence of a second, unrelated peptide carried into a batch from shared synthesis, purification or filling equipment, and it is one of the few quality failures a certificate of analysis is structurally unable to detect. A chromatographic purity figure derived from ultraviolet area percent will often report a contaminated batch as 99 percent pure, because the contaminating peptide either co-elutes with the main peak or absorbs too weakly to register. Documented analysis of grey market peptide preparations has confirmed this failure mode in practice.

The distinction matters because purity and identity are separate claims resolved by separate methods. A purity number answers the question of how much of the ultraviolet signal belongs to the largest peak. It does not answer the question of what else is in the vial. When a manufacturer runs several different peptides through the same reactor, the same preparative column and the same lyophilizer, the residue left behind by the previous campaign becomes an impurity in the next one. That residue is chemically unrelated to the labelled compound, so none of the usual synthesis impurity logic applies to it, and the specification on the certificate almost never names it.

Why a Purity Percentage Cannot See a Second Peptide

Reversed phase high performance liquid chromatography with ultraviolet detection is the workhorse of peptide release testing. Detection at 214 nanometres targets the amide backbone, and detection at 280 nanometres targets aromatic residues. Both are indirect. The signal a compound produces depends on its extinction coefficient, which varies substantially with sequence composition. A short peptide with no tryptophan, tyrosine or phenylalanine produces a weak response at 280 nanometres and a modest one at 214 nanometres relative to its actual mass in the vial.

This creates a specific blind spot. If a contaminating peptide is present at a low percentage by mass and has a lower extinction coefficient than the labelled compound, its contribution to total peak area shrinks below the reporting threshold of the method. It is present, it is quantifiable by mass spectrometry, and it never appears on the certificate. The same problem is discussed in more general form in our analysis of what a certificate purity number leaves undefined, where area percent is shown to assume that every species in the sample absorbs light equally.

Co-elution compounds the problem. Two peptides of similar hydrophobicity will emerge from a C18 column at nearly the same retention time. Integrated as a single peak, they produce one purity figure that flatters both. Only an orthogonal method with mass resolution separates them.

The Documented Case: AOD-9604 Where It Should Not Have Been

The most direct published evidence for this failure mode comes from Janvier and colleagues, writing in Talanta in 2018 (volume 188, pages 795 to 807). The group performed a systematic impurity screen on the ten most frequently encountered falsified peptide preparations on the Belgian market, sourced from three separate illegal internet vendors that sold their material under research chemical framing. Twenty seven samples were characterised for active ingredient content, sequence related impurities, small molecule contaminants, elemental impurities and residual solvents.

Buried in the results is a finding that deserves far more attention than it has received. The authors report that contamination with a second active ingredient, AOD-9604, was identified by ion trap mass spectrometry in the GHRP-2 preparation from one vendor and the sermorelin preparation from another. They state explicitly that this contaminating peptide was not detected in the corresponding ultraviolet chromatograms. Two products from two different vendors carried a peptide that had nothing to do with the labelled compound, and the detection method that generates the purity number on a certificate registered nothing at all.

The purity figures for those same GHRP-2 preparations were 99.6, 99.4 and 99.9 percent across the three vendors. A researcher reading a certificate reporting 99.4 percent would have no reason to suspect that a growth hormone fragment analogue was also in the vial. AOD-9604 is a distinct compound with a distinct receptor profile, and its unlabelled presence in a secretagogue preparation invalidates any experiment that assumes a single variable.

The same study documented a related identity failure. Full length thymosin beta-4 was found in all three preparations examined, although two of the three labels claimed the shorter synthetic fragment marketed as TB-500. That is a substitution rather than a carryover, but the mechanism of discovery was identical: mass spectrometry caught what ultraviolet detection could not.

Key Research Findings

• Janvier et al., Talanta, 2018, 188:795-807, analysed 27 samples covering 10 peptide compounds from 3 illegal internet vendors.
• AOD-9604 was detected as a contaminating active ingredient in GHRP-2 from one vendor and sermorelin from another, identified only by LC ion trap mass spectrometry and absent from the ultraviolet chromatograms.
• Measured purity for cysteine containing peptides in that sample set ranged from 5 percent to 75 percent, with the three AOD-9604 preparations returning 45.9, 4.8 and 4.8 percent.
• Six of the samples exceeded the ICH class 1 elemental limit for arsenic of 1500 parts per billion, with measured concentrations from 1660 to 12,890 parts per billion, and speciation confirmed all arsenic was present in the more toxic inorganic form.
• Oxytocin preparations returned purities of 74.8 and 71.7 percent, with dimer and trimer species accounting for 10.5 and 13.0 percent relative peak area in a single sample.
• Fourman and Mullen, Pharmaceutical Technology, 1993, 17(4):54-60, established the two carryover criteria still in general use: 10 parts per million of a residual compound in the following product, and one thousandth of the smallest daily therapeutic quantity, whichever is more stringent.

How Cross-Contamination Enters a Peptide Batch

Solid phase peptide synthesis is a cyclical process run in a reaction vessel that is drained, washed and reloaded between campaigns. Resin fines, cleaved fragments and adsorbed material persist in vessel seals, frits, transfer lines and valve dead legs. Peptides are unusually prone to this because they adsorb tenaciously to stainless steel, polypropylene and glass, a behaviour we have examined in the context of how certificates are verified and where verification stops. Surface loss that costs a laboratory recovery in an assay is the same phenomenon that leaves residue in a production vessel.

Preparative chromatography is the higher risk step. A preparative C18 column represents a large capital cost, and running one column across multiple products is standard economics in contract peptide manufacturing. Strongly retained material from a previous injection can bleed into subsequent runs over many cycles. Analytical chemists know this as carryover, and bioanalytical method validation requires it to be measured for exactly this reason. At preparative scale the same physics operates on a larger mass basis.

The lyophilizer and the fill line close the loop. Freeze drying chambers are shared across products, shelves and condensers accumulate residue, and aerosolised powder during filling is a recognised vector for airborne cross-contamination between adjacent operations. A facility that runs 30 different peptides through one suite has 30 opportunities for the previous compound to appear in the next one.

The Limits That Govern Carryover

Regulated manufacturing addresses this through cleaning validation, which is documented evidence that a cleaning procedure reduces residue to a predetermined acceptable level. The foundational quantitative framework came from Fourman and Mullen in Pharmaceutical Technology in 1993, and it set two parallel criteria. The first caps a residual compound at 10 parts per million in the subsequently manufactured product. The second caps it at one thousandth of the smallest daily therapeutic quantity of that compound. The stricter of the two governs, and equipment must additionally be visually clean.

The European Medicines Agency replaced the blunt version of this arithmetic with a toxicology driven approach in EMA/CHMP/CVMP/SWP/169430/2012, the guideline on setting health based exposure limits for products manufactured in shared facilities. Adopted by the European Medicines Agency human medicines committee on 20 November 2014 and effective from 1 June 2015, it requires a permitted daily exposure value derived from the toxicological profile of each compound rather than a single generic threshold. For highly sensitising or highly potent compounds, the guideline pushes toward dedicated equipment because no achievable cleaning limit is defensible.

Peptides sit awkwardly in this framework. Many research peptides have no established toxicological profile from which to derive a permitted daily exposure value, because they have never completed the preclinical package that would generate one. A manufacturer that wanted to set a defensible carryover limit for a compound like AOD-9604 into a compound like sermorelin would find the underlying toxicology data does not exist.

The Canadian Regulatory Position

Health Canada addresses cleaning validation in GUI-0028, the cleaning validation guide, most recently revised on 29 June 2021, which supports Part C Division 2 of the Food and Drug Regulations. The guide covers equipment qualification, analytical method suitability, sampling by swab and rinse, establishment of residue limits and change control. Its position on highly sensitising compounds mirrors the European one: where limits fall below the detection capability of the best available analytical method, dedicated facilities are the practical answer.

ICH Q7, the good manufacturing practice guide for active pharmaceutical ingredients, carries parallel expectations in its cleaning validation provisions. None of these frameworks bind a supplier selling material for research purposes only, which is precisely the gap the Janvier data illustrates. The frameworks exist and are well specified. They simply do not reach the grey market, and a certificate issued by a contract laboratory testing a finished vial says nothing about whether the facility that produced it operated any cleaning validation programme at all.

What This Means When Reading a Certificate

A batch specific certificate that reports high performance liquid chromatography purity and mass spectrometry identity is meaningfully better than one that reports purity alone, because mass spectrometry is the method that caught the AOD-9604 contamination the ultraviolet trace missed. The relevant question is what the mass spectrometry section actually reports. A single confirmed molecular ion for the labelled compound establishes that the compound is present. It does not establish that nothing else is. A full scan chromatogram with the identity of secondary peaks assigned is a stronger document than a one line molecular weight confirmation.

The second question concerns facility practice rather than analysis. A supplier that runs a dedicated production line for a given compound, or that can describe its changeover procedure between campaigns, is making a claim that can be examined. A supplier that cannot answer the question is not necessarily contaminating material, but the risk is unquantified. Our full set of batch documents is published on the certificates of analysis page, and the compounds those documents cover are listed across the research catalogue.

The third point is methodological rather than commercial. Any experiment that produces an unexpected result with grey market material has an unexamined confounder available. Before a novel receptor interaction is attributed to the labelled compound, the possibility that a second peptide is present at a few percent by mass deserves elimination by orthogonal analysis of the actual vial used.

Frequently Asked Questions

Can a peptide be 99 percent pure and still contain another peptide?

Yes. Purity by ultraviolet area percent measures the fraction of total absorbance belonging to the main peak. A contaminating peptide that co-elutes with the main peak, or that has a substantially lower extinction coefficient, contributes little or no distinguishable area. Janvier and colleagues documented exactly this in preparations reporting purities above 99 percent.

Which analytical method detects cross-contamination?

Liquid chromatography coupled to mass spectrometry. Mass resolution separates species that co-elute chromatographically, and full scan acquisition surveys the sample for masses that were not expected rather than confirming only the mass that was. Ultraviolet detection alone cannot perform this function regardless of column quality or run length.

Does a third party certificate rule out cross-contamination?

Not by itself. An independent laboratory tests the sample it receives against the specification it is asked to apply. If the specification requests purity and identity confirmation for one compound, the laboratory reports on that compound. Detecting an unlabelled second peptide requires the analytical scope to include untargeted screening, which is a separate and more expensive request.

Why does peptide manufacturing carry higher carryover risk than small molecule manufacturing?

Peptides adsorb strongly to the surfaces of processing equipment, they are frequently produced in multi product facilities on shared preparative chromatography systems, and many research peptides lack the toxicological data required to set a health based exposure limit for cleaning validation purposes. The combination of high surface affinity and absent limit setting data is unusual.

Summary

Peptide cross-contamination is a documented, published failure mode in grey market peptide supply, and it is invisible to the analytical method that produces the purity figure most buyers rely on. The regulatory frameworks that address it, from the Fourman and Mullen carryover criteria through the European health based exposure limit guideline to Health Canada GUI-0028, are mature and specific. They do not automatically apply to material sold for research purposes only. The practical response is to read the mass spectrometry section of a certificate as carefully as the purity line, and to treat an unexplained result from unverified material as an analytical question before it becomes a scientific one.

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