Considerations for Pre-specifying the Change Categories for Manufacturing Process Parameters Identified as Established Conditions for Drug Products (Chemical Products) — PMDA Early Consideration Summary

In Plain Language

The document considers how a sponsor can set in advance the regulatory change category for a manufacturing process parameter identified as an Established Condition (EC). Its central question is how the parameter’s Proven Acceptable Range (PAR) relates to the range or value proposed for the application and to normal manufacturing operation. The document distinguishes situations in which a parameter is critical to product quality from situations in which it is not classified as critical, and it treats proximity to the edge of the PAR as important. If the proposed description is the PAR itself, or a range close to its limits, the document says a partial change approval application category must be used; a range sufficiently inside the PAR may be designated for minor change notification, with a scientific justification. This approach is intended to take account of the potential risk to product quality and the evidence that actual manufacturing variability remains within an acceptable range.

What This Guidance Covers

This Early Consideration document addresses drug products limited to chemical products. It concerns manufacturing process parameters that have been identified as ECs on the basis of the concepts in the ICH Q12 Guideline, and the advance designation of change categories for those parameters in Japan. In this context, the application’s Manufacturing Method section describes the parameters, ranges, or values under consideration. The categories discussed are items requiring a partial change approval application and items subject to minor change notification.

The document’s main focus is the relationship between PAR and the Normal Operating Range (NOR), which it describes as the range of normal operation or procedures specified in product standard specifications and similar documents. It sets out basic considerations rather than a single universal decision rule: the document says the approaches are examples and do not preclude other science- and risk-based approaches. It also states that the considerations may be referred to, as appropriate, even when the ECs were not identified using the ICH Q12 concepts.

Several matters are expressly outside scope. The document does not address whether process parameters should be described in the application as target values or set values. It also excludes cases where PAR evaluation is insufficient—for example, where only a single plot under one specific condition is presented instead of evaluating PAR as a range. The document does not necessarily guarantee that content described in a marketing application will be accepted.

Why It Matters

The document explains that, under the ICH Q12 approach, ECs are legally binding information considered necessary to assure product quality, and changes to ECs after approval require regulatory procedures. In Japan, manufacturing process parameters identified as ECs must be described in the Manufacturing Method section of the marketing application. The issue addressed here is how to decide, before a change occurs, which change category applies to a described parameter or range.

That decision needs to reflect the possibility that changing a process parameter could affect quality. A range that lies inside PAR may still be too close to its boundary to provide adequate assurance under actual manufacturing conditions. Conversely, a range with a supported margin from the PAR limits may be considered for minor change notification, depending on the relevant evidence and risks. The document therefore connects the proposed application description to process understanding, equipment performance, operating variability, and the potential consequences of a process failure.

Key Recommendations

1. Assess risk and define what will be described. When setting a category in advance for a process parameter identified as an EC, it is necessary to consider potential risk to product quality and determine both the range or value to be described in the Manufacturing Method section and its change category. Factors that may be considered include process understanding and robustness, equipment characteristics, difficulty of control, PAR, NOR, and the edge of failure—the boundary beyond which relevant quality attributes can no longer meet acceptable criteria.

2. Account for severity even when NOR is inside PAR. The document says category decisions are fundamentally based on potential risk assessment, described as severity of harm multiplied by probability of occurrence of harm, in accordance with the ICH Q9 Guideline. Even if NOR is sufficiently within PAR, where process failure could cause extremely serious harm—for example, for parameters affecting quality attributes related to sterility or viral inactivation—it is appropriate to designate the change conservatively as requiring a partial change approval application.

3. When the edge of failure is at the PAR limit, do not allow deviation from PAR. The document describes PAR, in general, as the range for a single parameter within which acceptable quality meeting relevant specifications can be obtained when other parameters are held constant. Where the edge of failure corresponds to the PAR limits, deviation from PAR is not acceptable. The application may describe PAR itself or a range or value within PAR; the choice should be determined case by case, taking actual manufacturing control into account.

4. For a CPP, treat an unidentified edge of failure cautiously. A Critical Process Parameter (CPP) is one whose variability affects a Critical Quality Attribute (CQA) and therefore should be monitored or controlled to ensure the desired quality. Even if an edge of failure has not been identified within the range evaluated, the document says it is reasonable to consider that it exists relatively close to or just outside the PAR limits. Because an effect outside PAR cannot be excluded, deviation from PAR is not acceptable in this situation too; the parameter should be described in the application using the same approach as when the edge of failure is at the PAR limit.

5. Choose the category according to the described range’s margin from PAR limits. If PAR itself is described in the application, the item must be designated as requiring a partial change approval application, and the application must explain in CTD Modules 2.3 and 3 that deviation from PAR will not occur during normal operation, taking NOR and the relevant equipment and other-parameter variation into account. If a narrower range or value within PAR is described but sufficient margin from PAR limits cannot be secured, it must also be designated in that category. If the range or value is sufficiently within PAR, including consideration of equipment performance and operational variability, it may be designated as subject to minor change notification.

6. Support a minor change notification designation with evidence and an explanation. For a range or value sufficiently within PAR, the appropriateness of the minor change notification category must be described and justified in CTD Modules 2.3 and 3, or CTD Module 1.13. The justification is to be based on PAR-related study results, the range or value described in the application, and the performance and operational variability of the equipment used. The document says it is desirable not to rely only on qualitative claims such as “sufficiently controllable by the equipment used,” but to provide objective evidence demonstrating scientific validity.

7. For a non-CPP EC with no identified edge of failure, explain the basis for the category. If the parameter is not a CPP, its edge of failure has not been identified, and the possibility of an effect on product quality cannot reasonably be excluded, PAR or a range or value within PAR may be described as subject to minor change notification. The appropriateness of determining that the parameter is not a CPP and assigning this category must be explained in CTD Modules 2.3 and 3, or CTD Module 1.13, taking account of the parameter’s impact on CQAs and PAR study results.

8. Re-evaluate categories over the product lifecycle. Categories set after approval are not fixed. The document says they should be continuously re-evaluated as manufacturing conditions and knowledge change. For example, adding manufacturing sites, changing equipment, or accumulating long-term manufacturing experience may change the initial margin from PAR limits or the process risk profile. If sufficient margin can no longer be secured in a new manufacturing environment, the category and parameter range or value should be appropriately reviewed through change control procedures.

Practical Implications

Interpretation: In practice, sponsors need to connect the wording proposed for the application to the evidence about the parameter and the process as actually operated. A category rationale is not just a label: for a proposed minor change notification category, the application explanation should make clear how the proposed range sits in relation to PAR and why equipment performance and operational variability do not remove the margin relied on.

Interpretation: Quality and process-development teams can contribute evidence about PAR studies, process validation, equipment calibration and validation, manufacturing experience, and process capability. The document identifies process capability indices (Cp/Cpk) based on manufacturing experience using the actual equipment, process validation results, and equipment calibration and validation results as examples of objective evidence that may be used. These are examples, not a stated exhaustive list or a universal set of required data.

Interpretation: Regulatory teams and statisticians should ensure that the justification matches the proposed description and the intended category. Where interactions among parameters are important, assessing each parameter independently may not be appropriate from a quality-assurance perspective. The document points to multivariate analysis, such as Design of Experiments (DoE), as a way to explain that a PAR setting remains appropriate within allowable variation of other parameters. Where interaction effects are significant and a single-parameter range cannot adequately define PAR, it says to consider multivariate control, such as a model or design space.

Interpretation: Manufacturing and lifecycle teams should not assume that a category established at initial approval will remain suitable indefinitely. Site additions, equipment changes, and accumulated experience can alter the operational picture. A change-control process can provide a route to revisit the category and the described range when the original margin or risk assessment no longer fits the manufacturing environment.

Technical Details

The document uses the edge of failure to frame the relationship between process variation and quality. It is the boundary beyond which relevant quality attributes can no longer meet acceptable criteria. The decision logic differs according to whether that boundary is identified at the PAR limit, has not been identified for a CPP, or has not been identified for a parameter that is not a CPP.

Where multiple parameters interact significantly, determining a separate PAR for each parameter may not adequately assure quality. The document says to consider explaining the appropriateness of PAR under allowable variation of other parameters using multivariate analysis such as DoE. If interactions are significant and single-parameter PAR determination is difficult, sponsors should consider describing control using a model or design space. The change category for a model or design space is determined case by case according to risk associated with changes to the model; the document notes that a specified section of the ICH Quality Implementation Working Group Points to Consider (R2) may also be referenced when considering that risk.

The document does not set numerical margins, Cp/Cpk cut-offs, or a universal formula for translating a margin into a change category. Its framework is based on the case-specific relationship among PAR, the proposed description, operational variability, and potential quality risk. In particular, potential risk assessment considers both severity and probability; a low probability alone does not remove the need for conservative categorisation when the potential harm is extremely high.

Who Should Read It

This document is relevant to regulatory affairs professionals preparing or reviewing Japanese marketing applications, especially the Manufacturing Method section and its change-category descriptions. It will also be useful to quality and CMC professionals, process-development scientists, manufacturing and validation teams, statisticians supporting PAR or multivariate analyses, and lifecycle-management teams considering how changes to sites, equipment, or process experience affect an established category. Investigators and clinical operations staff may find it relevant when they need to understand how manufacturing process control and product-quality considerations are represented in application materials.

Official PMDA Source

• Official title: Considerations for Pre-specifying the Change Categories for Manufacturing Process Parameters Identified as Established Conditions for Drug Products (Chemical Products) (Early Consideration)
• Issuing body: Pharmaceuticals and Medical Devices Agency (PMDA), Center for Product Evaluation
• Date printed on the document: March 12, 2026
• PMDA PDF: Official PMDA PDF
• PMDA list page: PMDA Early Consideration documents

Source: Pharmaceuticals and Medical Devices Agency (PMDA) website, PMDA Early Consideration documents. This page was created by editing and summarising the PMDA document.

Consult the official document for its full context; this summary does not replace it. PMDA’s English version is provided for reference, and the Japanese original prevails if there is any inconsistency. PMDA describes Early Consideration documents as reference information reflecting its views at the time of publication; they may change as scientific knowledge develops.

Related Clinical Biostats Resources

• All Clinical Biostats tutorials — Browse other plain-language tutorials across clinical research and regulatory topics.
• Statistical Calculators — Access calculators that may support quantitative work in research and analysis.