Matrix: Cleaning Validation Worst-Case Product Grouping and Selection

This is a ready-to-use worst-case grouping matrix for cleaning validation on a shared equipment train. It is the living document that decides which product-to-product transition governs the cleaning limit, so validation is run against the hardest case rather than a convenient one. Replace every <<FILL: ...>> placeholder with your own specifics and keep it under change control. A worked specimen follows. Verify each cited regulation against the current source before you rely on it.

Document control header

1. Purpose

Identify, for <<FILL: EQUIPMENT TRAIN>>, the worst-case product transition that governs the cleaning acceptance limit, so cleaning validation demonstrates control of the hardest case and, by bracketing, all easier transitions. The matrix also defines the product grouping (family/bracketing) rationale and the trigger to re-select when the product mix changes.

2. Methodology

Worst case is selected on the combination of factors that make a residue both hazardous and hard to remove and detect. For each product as a potential contaminant (product A) into each next product (product B), the governing case is the one producing the lowest maximum allowable carryover (MACO) and the greatest cleaning difficulty.

Selection factors:

• Toxicity / potency: the PDE from the signed HBEL monograph. A low PDE lowers MACO.
• Cleanability / solubility: how hard the residue is to remove (poor solubility, adhesion, drying, degradation to hard-to-clean species).
• Dose and batch size of the next product: a small batch size or a high maximum daily dose of product B lowers MACO.
• Detectability: whether the analytical method can quantify at the resulting limit.

Bracketing/grouping is acceptable where products share equipment, cleaning procedure, and residue characteristics closely enough that validating the worst case represents the group. State the grouping rationale; grouping without a rationale is a finding.

3. Product property matrix

One row per product that runs on the train. Pull the PDE from each product’s HBEL monograph.

4. MACO transition matrix

For each contaminant A into each next product B, compute MACO = (PDE_A x MBS_B) / MDD_B. The lowest MACO governs. Mark the governing cell.

Governing (lowest) MACO: <<FILL: value>>, transition <<FILL: A into B>>.

Convert the governing MACO to the per-swab acceptance limit: divide by the shared surface area for the area-normalized limit, multiply by the swab area, then divide by the recovery factor. Record the final recovery-corrected limit the validation must meet.

• Area-normalized limit: <<FILL: µg/cm²>>
• Recovery-corrected swab limit: <<FILL: µg/swab>>
• Method LOQ below the limit with margin: <<FILL: yes/no>>

5. Worst-case cleaning-difficulty check

The lowest-MACO transition is usually the worst case, but confirm the hardest-to-clean product is captured. If the lowest-MACO product and the hardest-to-clean product differ, validate both, or justify why the governing case bounds the other.

• Lowest-MACO product: <<FILL>>
• Hardest-to-clean product (cleanability rank): <<FILL>>
• Both captured by the validation? <<FILL: yes; or justify>>

6. Acceptance criteria

• Every product on the train has a row and a PDE from a signed HBEL monograph.
• The MACO matrix is complete for all credible transitions and the governing cell is marked.
• The governing MACO converts to a recovery-corrected limit the method can detect with margin.
• The hardest-to-clean product is captured by, or bounded by, the validated case.
• The grouping/bracketing rationale is documented.
• Change control re-triggers the matrix on any new product, formula, equipment, or HBEL change.

7. References

EMA/CHMP/CVMP/SWP/246844/2018 (Q&A on risk-based cross-contamination) for the HBEL-to-limit flow. EU GMP Chapters 3 and 5 (2015 revision). ICH Q9(R1), Quality Risk Management, for the grouping and worst-case rationale. PIC/S recommendations on cleaning validation (reference by number/title; describe, do not paste).

Confirm the current version and clause numbers of each reference before issue.

8. Revision history

9. Approvals

Filled specimen

The following shows the matrix completed for three illustrative products on one solid-dose train (shared surface area 25,000 cm², swab area 25 cm², recovery 80%). Numbers are illustrative.

Product properties:

MACO (mg of A across the whole batch of B), lowest cell in bold logic:

Governing MACO: 2,500 mg, Bravo into Alfa or Bravo into Charlie (Bravo is the potent contaminant, PDE 0.05 mg/day). Converting 2,500 mg over 25,000 cm² gives 100 µg/cm²; times a 25 cm² swab is 2,500 µg/swab; divided by 0.80 recovery is a 3,125 µg/swab acceptance limit. Bravo is also the second-hardest to clean, so the validation targets Bravo removal at the worst-case transition. If a new, more potent product were added, the matrix would be recomputed and the governing cell could flip.

Common inspection findings this matrix prevents

• Validation run against an easy transition while the potent, low-MACO product was never the target.
• A worst case asserted with no calculation behind it.
• The hardest-to-clean product ignored because a different product had the lowest MACO.
• Bracketing/grouping used with no documented rationale.
• A new product added to the train with no recomputation, leaving a stale worst case.

How to adapt this matrix

1. List every product on the train and pull each PDE from its signed HBEL monograph.
2. Compute MACO for every credible A-into-B transition and mark the lowest.
3. Convert the governing MACO to a recovery-corrected swab limit and confirm the method LOQ sits below it.
4. Check the hardest-to-clean product is captured; validate both cases if they differ.
5. Document the grouping rationale and wire change control to recompute on any new product or new HBEL.