Cross-contamination in peptide plants

Most peptide factories make more than one product, and many make dozens on the same equipment. That is normal and not automatically a problem. The problem is the assumption that “same line” means “same cleanliness” — a claim that has to be demonstrated lot by lot, not inferred from a certificate.

Three different failures

The words are often used interchangeably, which makes the corrective action unclear:

Failure	What happened	Typical evidence
Mix-up	A different product or lot was packed, labelled or shipped as the ordered one	Wrong label, wrong vial, identity test mismatch
Cross-contamination	Material from another product is physically present in this lot	An extra peak or mass not attributable to the molecule; foreign morphology
Carryover	Residue from a previous batch enters the next one through shared equipment	Trace signal that falls with successive batches; cleaning records

They have different controls. Mix-up is prevented by line clearance and labelling; cross-contamination by containment and cleaning; carryover by validated changeover. A supplier who answers all three with “we follow GMP” has not answered any of them.

Where material travels

In peptide production the shared surfaces are numerous, and several are easy to overlook because they are not the reactor:

Reaction and cleavage vessels. The most obvious shared surface; also the easiest to clean and verify.
Filtration and transfer lines. Tubing, valves, sight glasses and pump heads retain liquid and are hard to swab.
Purification columns. A column used for several products can carry previous material in the bed, frits or dead volume. Resin reuse is a change-control question in its own right.
Lyophiliser shelves and trays. Dried powder is light, electrostatic and easily airborne during unloading.
Milling, sieving and blending. Dry operations generate dust that settles on surfaces outside the immediate equipment.
Filling and capping. Product contact parts, bowls, augers and the surrounding bench.
HVAC and room surfaces. Airborne powder can migrate; cleaning of walls, floors and ceilings is part of changeover, not housekeeping.
People and shared tools. Gloves, scoops, spatulas, balances and notebooks move between areas more freely than material does.

For a lyophilised product, the dry stages deserve more attention than the solution stages. A spill of solution is visible and local; a puff of powder during tray unloading is neither.

What cleaning verification can show

“Cleaned” is an observation; “verified clean” is a measurement with a stated method and limit. The useful distinction is between three levels:

Visually clean. No residue visible under defined lighting. Necessary, cheap, and blind to the amounts that matter for a potent peptide.
Specific marker. A known worst-case compound from the previous product measured by a specific method — HPLC, LC-MS or a validated UV assay — on a swab or rinse sample.
Non-specific total. TOC or conductivity of the final rinse, which detects general residue but cannot tell you which product it came from.

The three are complementary. Visual inspection catches gross failure; the specific marker proves the previous product is below the limit; the non-specific total catches residues the marker method does not see. A cleaning validation that reports only “visually clean” has not measured carryover.

Two details decide whether the result means anything. First, the recovery from the swab or rinse must be established — a method that recovers 30% of the residue and applies a 100% assumption understates the contamination. Second, the acceptance limit must be justified for the product and the route of exposure. For research material there is no universal health-based limit to borrow; the limit has to be agreed in the specification or contract, and it should be stated in the same units as the result.

Changeover is a procedure, not a wipe

A changeover between products typically includes line clearance, dismantling, cleaning, drying, inspection and re-assembly, with a record at each step. The questions that reveal whether the procedure is real:

Is the changeover procedure product-pair specific, or one generic SOP for every combination?
Which product is treated as the worst case, and why?
Are dedicated parts used for any product, and which ones?
How is drying verified before the next batch? Residual water can carry residue and support microbial growth.
Who signs the line-clearance record, and can the record be linked to the batch that follows?
What happens after a spill, a filter burst or a deviation — is the line re-cleaned and re-verified?

What the certificate cannot tell you

A COA reports the result for the lot in front of you. It does not report what else was in the facility, what ran before it, or whether the cleaning was verified. Those facts live in the batch record and the changeover record, which is why a documentation request is part of supplier qualification rather than an optional extra.

Analytically, low-level cross-contamination is hard to see. An HPLC method optimised for the main peak may not retain or resolve a contaminant from a different peptide; a mass spec method may not be run at all. A clean chromatogram is therefore compatible with a small amount of another product. The absence of evidence is not evidence of absence, and the supplier's changeover record is the evidence that fills the gap.

Practical questions before sharing a line

Question	Why it matters
Which products run on this line, and in what sequence?	Reveals the worst-case neighbour and the cleaning challenge
Is the line dedicated, campaign-based or fully multi-product?	Determines how often changeover evidence should exist
What was the last product before my lot?	Identifies the contaminant you would need to look for
Can I see the changeover record for my batch?	Turns a claim into a document
How is cleaning verified — visual, marker, TOC or all three?	Distinguishes a wipe from a validated procedure
What is the acceptance limit and who set it?	A result without a limit cannot be judged
Are columns and resin shared or dedicated?	Shared beds are a less obvious carryover route
What happens to the first lot after a campaign change?	First-lot testing is a common risk control

Red flags

“We only make this product” contradicted by the catalogue or the site tour.
Cleaning records that are identical across products and dates.
No stated acceptance limit, or a limit expressed only as “pass/fail”.
A refusal to name the previous product on the line.
Shared columns with no resin-life or carryover data.
Changeover records that cannot be linked to the batch record.

None of these proves contamination. Each one means the supplier cannot demonstrate control, which for a buyer is the same practical problem.

For laboratory research use only. This article describes facility and changeover practice; it makes no claim about the cleanliness of any specific supplier or product and does not support administration to humans or animals.

Editor's take

Cross-contamination is a line-level failure that looks like a product-level mystery. When a trace impurity appears in one compound but not its neighbours, suspect the shared equipment before the synthesis.

Limitations

This guide describes methods and interpretation, not acceptance limits. Limits must come from your own specification and, where applicable, the relevant pharmacopoeial monograph. No testing was performed by the author for this page.