Introduction
This document outlines the considerations site quality organizations should review to assess whether stability data are needed to support planned or unplanned post-approval changes to API or API Intermediate for sale manufacturing processes. This document also recommends customer notification if subsequent intermediate processing or drug product quality may be impacted by the change. This guidance also details considerations for the design of stability studies and the manner in which the resulting data should be used.
ICH Q7A states, “The potential for critical changes to affect established retest or expiry dates should be evaluated. If necessary, samples of the intermediate or API produced by the modified process can be placed on an accelerated stability program and/or can be added to the stability monitoring program.”
If a planned or unplanned process change impacts the registration, validation, or quality attribute(s) of an API, then the impact of this change on API stability should be evaluated for the following issues and documented.
- Significance of Process Change
- Stability Impact Assessment
- Data Analysis
- Stability Study Considerations
- Use of Stability Data
Recommendations & Rationale
1. Significance of Process Changes
ICH Q7A states, “Changes can be classified (e.g. as minor or major) depending on the nature and extent of the changes and the effects these changes may impart on the process.” A minor process change may be defined as a change that has no impact on registration or validation, and little or no impact on product quality. A minor process change should not require major testing or development efforts such as generation of supportive stability data. If stability data are needed, then the change should be categorized as major.
A major process change may be defined as a change that has a potential impact on registration, validation, product quality or GMP compliance. The evaluation of a major process change may require development support, including the collection of stability data.
In addition to determining whether a change is major or minor, the site Validation Team is responsible for determining whether stability data are needed to support a process change.
Examples of a major change could include, but are not limited to, the following:
- Process scale-up
- Implementation of a revised process
- Implementation of a new rework procedure
- Change in packaging or packaging supplier
- Change in storage conditions
- Change to existing process equipment (i.e. repair, modification, replacement not like for like)
- Transfer of the process to a new or different site
- Change in critical raw material production process or supplier (2)
2. Stability Impact Assessment
2.1. Regulatory Process Description Impact Assessment
If a change has the potential to impact the regulatory process description, then the impact on API stability should be evaluated and documented. The guidance document for “Product Change Management,” defines the minimum supporting documentation requirements for PCP/PCR submission based on the type of change. For example: a site change, a change to the primary package or a change in the storage conditions may require a stability study. Manufacturing Compliance should be consulted whenever there is a question regarding whether a change requires development support.
2.2. Validation Impact Assessment
If a change has the potential to impact validation, then the impact on stability should be evaluated and documented. At least one validation batch shall be placed on stability for processes that are new to a site or where it is believed that the change may affect stability, such as a new reprocess or rework step. All changes that may affect product quality or reproducibility of the process shall be evaluated for likely impact on the process or product. The determination of major or minor changes, need for and extent of stability testing shall be determined by the Site Validation Team (SVC).”
2.3. Chemical or Physical Attribute Impact Assessment
If a change has the potential to affect the chemical or physical attributes of an API or API Intermediate for sale, then the impact on stability should be evaluated and documented. If the affect of an Intermediate process change has potential to carry downstream to the attributes of the API or API Intermediate for Sale, then the impact of the change on stability should also be evaluated and documented.
ICH Q7A also states that internal and external drug product customers should be notified of changes that can impact API quality. Some common chemical and physical attributes which may be affected by process changes are, but are not limited to, the following:
- Particle size or distribution
- Polymorphic form (e.g. crystal form)
- Impurity profile
- Microbiological attributes
- pH
- Moisture Content (e.g. LOD, KF, Residual Solvents)
- Appearance and/or Color
3. Data Analysis
Assessing the impact of change on the stability of API or API Intermediates for Sale may include but is not limited to, the following steps:
Step 1: Comparison to registered specifications – Release results for all affected batches should meet registered specifications. In addition, no new impurities should be present in an amount greater than 0.1%. If all release requirements are met, then proceed to the next step where the release data are compared to historical batch release data to check for any atypical trends.
Step 2: Comparison to historical release data – Historical release data from the most recent batches prepared by the process prior to the change should be used as the “control” for data comparison purposes (lots involved with a previous process deviation may be excluded). It is desirable to have thirty or more batches in the historical data set. Fewer than 10 batches can be used for historical comparison, but the utility of some statistical tools is diminished. Data should be analyzed for any significant changes (i.e. data from affected batches is outside historical maximum, minimum, or average values), unusual trends, or new impurities. If release data for the affected batches still meet registered specification but an unexpected shift in data is noted, supportive stability studies may be needed.
Step 3: Analysis of existing stability data – Review of existing stability data will identify which chemical or physical attributes of an API or API Intermediate for Sale have been known to change over time. If the planned or unplanned change has potential to affect any of these attributes, then supportive stability studies may be needed.
4. Stability Study Considerations
Once it has been determined that a stability study is necessary to support a planned or unplanned change, the purpose of the study should be documented. The stability protocol does not need to be the same as for annual stability lots (e.g. accelerated or long-term conditions, study duration, test intervals, and test methods). The number of lots in the study will also be based upon the extent of the change. Stability data may be used to support a registration change, validation and/or batch release.
4.1. Change in Regulatory Process Description
If a process change results in a revision of the registered process description, supporting stability data may be required. Manufacturing Compliance Team with assistance from Quality Assurance and Regulatory Affairs, is responsible for the regulatory strategy and creation of the regulatory change packages and development of stability protocol.
European Medicines Agency (EMEA) requirements base stability protocol design on the type of process change.
A Type I Variation only requires supporting stability data for changes in API retest date or storage conditions. In either of these cases, stability data at label storage conditions are required for two batches covering the duration of the requested retest period.
A Type II Variation refers to changes that affect API attributes that have an impact on stability. For changes of this type, accelerated and long-term stability data are required. If the API is known to be unstable (less than 24 month retest interval), six months of data are required for three batches. If the API is known to be stable, then three months of data are required. In addition, if the change may impact the stability of the finished drug product, then drug product stability may also be required.
4.2. Validation
If a change requires validation, “One validation batch must be placed on stability regardless of whether or not it is a batch designated for commercial distribution and whether or not stability information is required for a regulatory filing. The number of batches to be placed on stability and the type of study required shall be documented and justified.”
Note that local regulations or commitments may require that the first three validation lots be placed on stability – these commitments take precedence over this guidance requirement.
4.3. Assignment of Retest or Expiry Dating for Affected Lots
ICH Q7A states that the potential for critical changes to affect retest or expiry dates should be evaluated. Accelerated stability conditions may be used to obtain data quickly (e.g. 3 months accelerated study) to support batch release for lots involved with planned or unplanned changes.
Without the support of these data, retest or expiry dating would be based only on available stability data from the new process – often resulting in shortened dating assignments until additional stability data are generated. If accelerated data are equivalent for lots produced before and after the change, then stability data from the original process can be used to support the retest or expiry dating for lots produced after the change.
5. Use of Stability Data
Stability studies can be initiated after batch release if the data will be used to fulfil annual stability commitments or if the data are for information only (e.g. analysis of a proposed package change or some other non-process related change). Batch release may need to wait until completion or partial completion of a stability study if the data are intended to (Note: this is not intended to be an all-inclusive list):
- Support a registered process change
- Support batch release
The amount of stability data required prior to batch release (e.g. completion of an accelerated study, 6 months of real-time data, etc.) should be determined and documented prior to initiation of the study.
Stability data generated for some deviation lots may only be intended to support batch release of the affected lots. Data from this kind of special stability study are not typically combined with data from annual stability lots to calculate retest or expiry dating for lots produced by the registered process. If stability data supporting a deviation are not consistent with historical stability data, then the affected batches may require rejection, rework or reprocessing prior to release.
References
If solvents are recovered on-site or off-site, the recovery process must be robust in separating other solvent components and solids from the recovery stream (e.g. multiple or packed distillation columns). It is expected that the recovered solvents have NMT 25ppm of dissolved solids to be considered to be animal-free or TSE risk-free. This requirement should be applied to solvents from any suppliers who use waste solvent as feedstock. If such suppliers are used, it is recommended that we understand if any of the waste solvent feedstock has had contact with animal and/or TSE risk materials.
References
– ICH Q7A Good Manufacturing Practices for Active Pharmaceutical Ingredients
– EMEA Stability Testing of Existing Active Substances and Related Finished Products (December 2003)
– EMEA Dossier Requirements for Type IA and TIB Notifications (July 2003)
– EMEA Stability Testing for Applications for Variations to a Marketing Authorization (May 2005)
– FDA Draft Guidance for Industry: Stability Testing of Drug Substances and Drug Products (June 1998)