Worst Case or Product-Specific? New Research Reopens the Cleaning Validation Strategy Debate

Headline

Worst Case or Product-Specific? New Research Reopens the Cleaning Validation Strategy Debate

Standfirst

A comparative review in the PDA Journal argues that neither worst-case matrices nor product-specific validation are universally correct — and positions visible residue limits as the risk-based bridge between them.

What happened

The authors report a structured comparison of the two dominant cleaning validation strategies, published ahead of print in the PDA Journal of Pharmaceutical Science and Technology on 18 April 2026. The paper, “Comparative Study of Matrix vs. Product-Specific Cleaning Validation Approaches” by Mahender Boyapally, carries DOI 10.5731/pdajpst.2026-000007.1 and PubMed PMID 42000198. It is a comparative review, not an experiment or simulation: the authors assess strengths, weaknesses, risk profiles and real-world applicability through regulatory guidance, industry best practice and case-study findings.

The authors report that the matrix (worst-case) approach — grouping related products and validating against a scientifically justified worst case, with acceptance limits typically derived from the most potent compound in the group — is highly resource-efficient, but shifts the burden onto the justification of the grouping and the worst-case selection itself. The product-specific approach validates every product transition individually, at a cost the authors characterise as enormous and, in large portfolios, difficult to sustain.

The authors report that the choice should be driven not by ideology but by three inputs: facility characteristics, portfolio complexity, and the formal risk assessment outcome. Their central conclusion is that a combined strategy — borrowing elements of both methodologies and using visible residue limits (VRL) as a risk-based instrument — may best balance regulatory compliance and operational performance.

Why it matters

Multi-product facilities have lived with an unresolved tension for years. Regulators expect a defensible scientific rationale for whatever scope is validated; operations expect that scope to stay affordable. A matrix approach leans on bracketing and worst-case rating so that only representative products are challenged, while group members are covered by argument rather than data. That argument is exactly where inspections concentrate: how groups were formed, how cleanability and solubility were compared, and whether the limit was set using the most potent compound in the group.

A product-specific programme removes those questions — at the price of a study per product transition, per equipment, per cleaning procedure. In a facility running dozens of products through shared trains, that growth makes the debate practical rather than academic.

What is particularly interesting

Three points stand out. First, the authors do not treat the two approaches as competing dogmas but as ends of a continuum, explicitly endorsing a combined strategy. Second, VRL is the pivot of that hybrid model: linked to the acceptable residue limit, it supports introducing new products into an already validated matrix without triggering a new validation campaign. Third, the framework is conditional rather than prescriptive — one facility may reasonably run matrix validation on one train and product-specific validation on another, if the risk assessment supports the division.

This is a deliberately unfashionable conclusion. Much recent discussion pushes toward either maximal worst-case conservatism or full product-by-product coverage. The authors argue that the middle ground, properly justified and supported by quantitative VRL data, is not a compromise on patient safety but an alignment of effort with actual risk.

Limitations

The paper is a review; the authors report no new experimental data, no simulated cleaning trials and no quantitative validation of the hybrid model. Its recommendations rest on interpretation of existing guidance, literature and case studies. It offers principles but no validated scoring system, acceptance criteria or worked example. VRL remains contested: acceptance of visual inspection as a surrogate for testing is conditional, and the risks of subjectivity and of a VRL sitting too close to the acceptable residue limit are real. The full text is paywalled, so the case-study analysis could not be independently examined.

Questions

• If we adopt a hybrid model, which shared equipment moves to product-specific validation, and what risk-assessment output justifies that split?
• How current is our worst-case rating, and how would a new product be screened against it before routine manufacture?
• Do we hold quantitative VRL data for our compounds, compared against the acceptable residue limit, established with defined viewing parameters?
• If a VRL introduces a new product without a new validation study, what evidence shows residues remain below the health-based limit?
• How would our validation master plan describe a combined strategy so it reads as one coherent, risk-based programme?
• Would our current scope survive a challenge that the worst-case selection was never formally risk-assessed?

Wider context

The paper lands in an established regulatory stream. ICH Q9(R1) names cleaning and validation as areas where quality risk management applies, and instructs that effort, formality and documentation be scaled to risk — while still requiring acceptable cleaning limits to be determined. The APIC cleaning validation guide has long endorsed bracketing and worst-case rating for multi-product API equipment, on the explicit condition that a documented scientific rationale for the chosen worst cases exists. Two decades of work on visible residue limits has shown that VRLs for most marketed formulations sit well below typical health-based limits, but also that regulatory acceptance is strongest where VRL use is anchored in an initial validation study with surface or rinsate testing.

Sources

• Boyapally M. Comparative Study of Matrix vs. Product-Specific Cleaning Validation Approaches. PDA J Pharm Sci Technol, 18 April 2026 (ahead of print). DOI 10.5731/pdajpst.2026-000007.1. PMID 42000198.
• ICH Q9(R1) Quality Risk Management, Step 4, January 2025.
• APIC, Cleaning Validation Guide, 2021.