A cleaning programme can pass its historical limits and still be poorly matched to today's product portfolio. A new compound may change the toxicological assessment, a smaller next batch may reduce the allowable carryover, or a longer campaign may create a residue that the original study never challenged. The central decision is whether the complete control strategy remains suitable for the actual sequence of products and equipment.
This article addresses cleaning strategy for shared and dedicated pharmaceutical equipment, linking health-based exposure limits, worst-case selection, validation and routine verification. It distinguishes patient exposure assessment from process capability and does not prescribe a universal carryover limit. The decision tools and examples below are GuideGxP recommendations that require site-specific scientific assessment and approval.
Define the contamination problem before calculating a limit
Map potential transfer pathways from the previous product to the next: retained material on contact surfaces, trapped liquid, common hoses, shared cleaning solution, portable parts and operator activities. Cleaning controls some pathways; segregation, containment, scheduling and handling controls address others. A successful residue result from one vessel cannot establish control of transfer through a shared room or an unassessed mobile component.
Distinguish product residues, cleaning-agent residues, degradation products and relevant microbial or endotoxin concerns. These hazards do not necessarily share an analytical method or acceptance rationale. A detergent that removes the product effectively may itself be difficult to rinse. A chemically clean but wet assembly may present a different microbial concern during storage. The strategy should identify the required equipment state at each handover.
Dedicated equipment changes the contamination scenario but does not eliminate cleaning requirements. Product degradation, residue accumulation, microbial proliferation and maintenance contaminants can remain relevant. Conversely, shared equipment is not acceptable merely because the proposed cleaning process looks technically strong. Suitability depends on the hazard, exposure pathways, achievable controls and applicable GMP provisions.
Regulatory framework and current status
EU GMP Chapters 3 and 5 address prevention of cross-contamination, while Annex 15, effective since October 2015, provides the cleaning validation framework. The EMA HBEL guideline, adopted in 2014 and effective in 2015, establishes the toxicological basis for assessing exposure to residues in shared facilities. The associated EMA questions and answers clarify its use within quality risk management. Current documents are available through EudraLex Volume 4 and the EMA GMP resource centre.
For US operations, applicable requirements include 21 CFR 211.67 on equipment cleaning and maintenance, with related controls for procedures and records. FDA's 1993 cleaning validation inspection guide remains useful historical inspection context; it is not a modern universal recipe for acceptance limits. The engineering decisions described here should be integrated with the site's pharmaceutical quality system and the applicable product and jurisdictional requirements.
HBEL and PDE: what they establish and what they do not
A health-based exposure limit describes an exposure level derived through a substance-specific scientific assessment. A permitted daily exposure is one form of HBEL. Its derivation requires appropriate toxicological expertise, relevant data, consideration of the critical effect and justified treatment of uncertainty. It is not generated by inserting the therapeutic dose into a convenient spreadsheet.
The toxicological report should identify the substance, assessment scope, route considerations, data sources, scientific reasoning, limitations and conditions requiring review. Where information is incomplete, the conclusion and associated uncertainty must remain visible to the team making the manufacturing decision. An unexplained number in a supplier file is inadequate input to a defensible cleaning strategy.
An HBEL informs a health-based carryover assessment; it does not become an instruction to operate near that exposure. Manufacturing controls should prevent uncontrolled contamination and reflect demonstrated process capability. Established cleaning limits may be more stringent for justified reasons, including the analytical strategy, process consistency or other product quality considerations. Any reassessment must explain both patient protection and continued control of the cleaning process.
Translate exposure assessment into a coherent acceptance strategy
A carryover calculation links the residue of a previous product to the quantity of a subsequent product that a patient could receive. The calculation depends on the relevant next product, batch assumptions, daily administration and transfer model. Record units and assumptions explicitly, including whether the entire shared equipment train or only part of it contributes to the assessed residue.
MACO is the maximum allowable carryover obtained from a defined calculation. It is one output within the strategy, not a substitute for it. Converting a train-level amount into surface or sample criteria introduces further assumptions about surface area, distribution, recovery and sampling coverage. The site must demonstrate that these transformations remain meaningful for its equipment and methods.
Do not apply 10 ppm, one thousandth of a dose or another legacy rule as a universal acceptance criterion. Historical criteria require scientific evaluation in the context of current knowledge and controls. Similarly, a method's quantification limit must not dictate an acceptable patient exposure simply because a more sensitive method would be inconvenient.
| Decision layer | Question to answer | Required input |
|---|---|---|
| Health-based assessment | What exposure assessment is appropriate for the substance? | Qualified toxicological evaluation and documented uncertainty |
| Carryover scenario | How could residue reach the next patient population? | Product sequence, equipment train and transfer assumptions |
| Sampling criterion | What result from this location or rinse supports the conclusion? | Area or volume, recovery, method capability and representativeness |
| Process control | What routine signals demonstrate execution within the established process? | Validated operating conditions, cycle records and verification plan |
Select worst cases across several dimensions
The most toxic product is not automatically the hardest to clean. A less hazardous formulation may adhere more strongly, dry more rapidly or contain excipients that obstruct removal. Evaluate the product and the formulated soil together. Solubility data should reflect relevant cleaning conditions; a room-temperature water value alone may poorly describe behaviour during the actual process.
Consider toxicological severity, potency where relevant, practical cleanability, formulation, residue loading, campaign duration, dirty hold time, heat history and analytical detectability. Assess the next product separately because its batch size and administration conditions can change the carryover scenario. Keep the direction of the transition clear: product A followed by B need not be equivalent to B followed by A.
Equipment selection needs the same care. Compare geometry, contact materials, seals, valve cavities, drainability, spray coverage, branch lengths and accessibility. A small vessel may be more challenging because its device arrangement creates a shadow; the largest vessel may be more challenging hydraulically. Selecting one by capacity alone can miss both effects.
| Candidate challenge | Why it may be limiting | Evidence needed before representation |
|---|---|---|
| Low-solubility product with adherent excipients | Physical removal may govern process development | Comparative cleaning development and relevant residue methods |
| Substance with a restrictive HBEL | Exposure assessment may govern acceptance | Toxicology, carryover assessment and analytical capability |
| Longest justified dirty hold | Drying or transformation may change the soil | Study conditions representative of actual storage and exposure |
| Complex valve or gasket interface | Delivery and sampling can be locally difficult | Design assessment, access demonstration and location-specific data |
Use bracketing and matrixing with explicit boundaries
A defensible grouping states what is shared and why the selected challenges represent the untested combinations. Relevant similarities may include product properties, cleaning chemistry, equipment geometry and operating ranges. Relevant differences must also be recorded. A scoring sheet can organise knowledge, but adding ordinal scores does not automatically establish scientific equivalence.
Define conditions that invalidate the grouping: a different soil mechanism, tighter health-based assessment, changed contact material, new gasket geometry, different spray arrangement or an extended hold time. Preserve the evidence linking each product and circuit to the representative study. If one challenge cannot represent both chemical difficulty and toxicological sensitivity, use more than one representative rather than forcing a single worst case.
Build development, validation and verification into one lifecycle
Development establishes how the cleaning process works and where its limits lie. It investigates chemical choice, mechanical action, temperature, contact time, rinsing and equipment configuration. Development trials can inform the operating envelope, but they do not become validation merely because their results are favourable. Validation follows an approved strategy with predefined acceptance criteria and justified challenge conditions.
Cleaning validation confirms the effectiveness of the defined procedure for its intended use. Routine cleaning verification checks specified evidence from actual execution, such as record review, visual examination and analytical sampling where required by the strategy. Verification supports continued assurance; repeated testing until a result passes does not repair an inadequately controlled process.
Separate dirty and clean hold times. For the former, consider soil drying, chemical change and microbial growth before cleaning. For the latter, consider drainage, drying, closure, storage conditions and subsequent handling. Campaign cleaning also requires a defined maximum operating pattern with its own scientific basis. Calendar duration alone can miss the influence of accumulated batches or interrupted production.
Practical example: introducing a new product into a shared train
A site operates a shared mixing vessel, transfer line and filling feed vessel. Its established matrix represents three readily soluble products. A proposed fourth product has a more restrictive toxicological assessment and a formulation that forms a persistent film after drying. The team identifies two independent challenges: the exposure scenario and the removal mechanism. Existing acceptable results do not resolve either by themselves.
The toxicologist reviews the substance and intended exposure context. Process development evaluates the dried film on relevant materials and checks detergent compatibility. Engineering investigates a valve cavity that is poorly reached by the current sequence. Analytical development demonstrates that the method can measure the intended residue at the proposed sampling criteria, including recovery from representative surfaces.
The project may retain the shared train, modify the cleaning procedure, replace a difficult component or reject the shared-use proposal. The decision follows the combined evidence. If a revised process is chosen, the site assesses its effect on the established products and cleaning-agent residues, updates the matrix and completes the required qualification or validation work before routine introduction.
Routine monitoring and change control
Trend more than pass or fail. Review shifts in rinse duration, detergent consumption, return behaviour, residue results, repeat cleaning, manual interventions and maintenance findings. A gradual increase in cleaning effort may reveal deteriorating delivery or changing soils before a formal acceptance failure. Interpret trends against changes in products and operating conditions rather than combining unlike circuits into one average.
Changes needing assessment include new products, suppliers or formulations; revised HBEL reports; altered batch sizes; longer campaigns; new detergents; equipment modifications; software changes; and new analytical methods. The review should identify affected assumptions and determine the necessary development, verification, requalification or revalidation. A change log that says only “no impact” without a technical rationale leaves the lifecycle disconnected.
Checklist, common mistakes and red flags
- Maintain a complete map of shared equipment and residue transfer pathways, including mobile and manually cleaned components.
- Keep the toxicological assessment distinct from the carryover calculation, sampling criterion and routine process endpoint.
- Justify both product and equipment representation, including untested combinations and their boundaries.
- Confirm analytical capability and recovery before relying on a sampling plan to support acceptance.
- Document campaign, dirty hold and clean hold conditions, with defined responses to excursions.
- Assign ownership for new product assessment, trend review and maintenance of the validation matrix.
Warning signs include using the lowest therapeutic dose as a complete hazard assessment, choosing only the largest vessel, treating an online conductivity result as proof of product residue removal, or declaring all products equivalent because the same detergent is used. Also investigate an increase in repeat cleaning, unexplained analytical variability and manual dismantling that is absent from the approved procedure.
Key takeaways and references
The strongest cleaning strategy connects hazard, transfer pathway, process capability and evidence. It uses HBEL expertise appropriately, selects worst cases by mechanism and maintains the reasoning as products and equipment change. Equipment design and sampling decisions should be developed together within the Cleaning, CIP & SIP Systems area; shared-room contamination interfaces may also require the Cleanrooms & HVAC Systems area.
- European Commission, EudraLex Volume 4: Chapters 3 and 5, Annex 15 and the linked EMA HBEL guideline.
- EMA good manufacturing practice resources: HBEL guidance and related questions and answers.
- 21 CFR 211.67: equipment cleaning and maintenance.
- FDA, Validation of Cleaning Processes, July 1993: historical inspection guidance.
- ICH Q9(R1) and Q10: quality risk management and pharmaceutical quality system context.
Related decisions
- URS for Cleaning, CIP & SIP Systems: Requirements, Structure and Checklist
- Cleaning Validation in GMP Manufacturing: Swab, Rinse, Recovery and Acceptance Strategy
Explore all decisions in Cleaning, CIP & SIP Systems.
