Should Cleaning Validation Extend Into the QC Laboratory?

Cleaning validation has historically been a production-floor discipline. A 2025 case study argues the same rigour belongs on the glassware and stainless steel of the QC laboratory — and offers a protocol for it.

What happened

The authors report a structured methodology for cleaning-validation protocols designed specifically for pharmaceutical quality-control laboratory equipment. Published in DARU Journal of Pharmaceutical Sciences (vol. 33, article 20, 2025; DOI 10.1007/s40199-025-00566-x; 3 July 2025) by Moura, Pereira, Santos, Silva, Paiva and Duarte.

It has three parts: identifying a worst-case API; recovery studies to optimise sampling method and solvent selection; and statistical refinement. Oxcarbazepine (Oxc) — described as poorly soluble and hard to clean — was the worst-case contaminant.

Validation covered 17 frequently used items spanning glassware and stainless steel. Two sampling methods (swabbing, rinsing) and two solvents (acetonitrile, acetone) were compared; each cleaning ran twice to test whether a second step was necessary, with four replicates per combination. API residues were quantified by HPLC on a Waters Alliance e2695 (Lichrospher 100 RP-8 column, 215 nm, 1.3 mL/min), with an LQ of 0.005 µg/mL, LD of 0.002 µg/mL, R² = 0.99997 and repeatability RSD 1.24 %.

Detergent residues were assessed separately: rinses from two specimens per equipment type were analysed by conductivity (Miron Ultrameter II 4P, 5 mL purified water) and TOC (Analytik Jena multi N/C 3100, 75 mL in nitric-acid-washed flasks). A company cleaning-agent limit of 10 ppm was applied — stricter than the European Pharmacopoeia purified-water limits referenced (4.3 µS/cm at 20 °C; 500 ppm TOC).

Why it matters

The authors report that swabbing achieved recovery above 70 % for both solvents after the first cleaning, with coefficients of variation below 10 %, and that rinsing attained recovery above 65 % on stainless steel. Current practice generally recommends a minimum recovery of 50 %; the tested combinations consistently exceeded that threshold. A second cleaning step may therefore be unnecessary here — though second-cleaning repeatability was poor across all conditions.

All conductivity measurements fell well below the 4.3 µS/cm limit, indicating negligible ionic contamination, and most TOC values sat below the analyser's quantification limit of 0.500 ppm — those above it remained under the 10 ppm acceptance threshold. Across the 17 items, the RSD of the Oxc signal did not exceed 2 %.

The practical weight of this paper is the blind spot it addresses. Cleaning-validation programmes routinely enumerate production vessels, filling lines and product-contact parts, while the volumetric flasks, dissolution vessels, pipettes and mortars holding reference standards and samples are often governed by general laboratory-hygiene SOPs rather than a validated protocol. The authors state plainly that case studies for laboratory equipment remain scarce.

What is particularly interesting

Reporting sampling recovery alongside cleaning effectiveness — rather than collapsing both into a single residue result — is the methodologically important move. Swab recovery above 70 % and rinse recovery above 65 % mean the downstream numbers represent the cleaning process, not the sampling method's losses.

Pairing a specific method (HPLC for the API) with non-specific ones (conductivity, TOC for the detergent) also reflects defensible practice: HPLC confirms removal of the known worst case, while the others catch surfactants and ionic residues a compound-specific assay would miss. Less common is the hypothesis test comparing average swab against average rinse recovery (H0: swab ≤ rinse; H1: swab > rinse) — a statistical justification for method selection.

Limitations

The authors are candid: validation used a single API and limited equipment surfaces under controlled conditions, and the work is potentially labour-intensive when extended. They recommend verification across APIs differing in physicochemical properties, potency and toxicological profile; a broader range of surfaces, geometries and hard-to-clean areas; variable cleaning conditions; adaptive risk-based strategies; more repetitions to capture batch-to-batch variability; and more sensitive techniques such as LC-MS/MS for degradation products and excipients. The second-cleaning finding is provisional given its poor repeatability. Implementation also requires adaptation to each company's context — a published protocol is a template, not a drop-in procedure.

Questions worth asking in our own programme

• Are our QC laboratory glassware and stainless-steel items covered by a validated cleaning protocol, or by general laboratory-hygiene instructions? Which items would qualify as our worst case?
• Have we performed recovery studies for our specific equipment classes, surfaces and sampling solvent/technique combinations?
• Is our sampling-method choice justified by data comparing swab and rinse recovery on the actual surface?
• Are cleaning-agent residues monitored by conductivity and TOC as well as by a compound-specific assay?
• What is our documented basis for the number of cleaning steps and sampling replicates, and have we shown a second cleaning adds value?
• Would our analytical methods detect degradation products and excipients, or only the parent API?

Wider context

Regulators have long expected a documented, scientifically justified basis for cleaning procedures; the study frames cleaning validation as controlling cross-contamination and keeping residues below maximum allowable carryover limits. Where the paper pushes is scope: the logic that applies to a production reactor applies to the volumetric flask in which a reference standard is prepared. For multi-product laboratories the question shifts from whether to extend, to where the risk ranking places the boundary — and what evidence justifies it. The approach reads as a template: pick the worst case, measure recovery honestly, separate API-specific from non-specific detection, and let the data set the protocol.

Sources

• Moura MJ, Pereira AD, Santos DJF, Silva AG, Paiva CCAD, Duarte BPM. Cleaning validation in pharmaceutical quality control laboratories: a structured protocol for contamination risk mitigation. DARU J Pharm Sci 33(2), 20 (2025). DOI: 10.1007/s40199-025-00566-x (PMID 40608165; PMC12229404)