| Title: Reduced Testing Program | |||||
| Guidance Number: 80 | |||||
| Prepared by: | Date: | Supersedes: | |||
| Checked by: | Date: | Date Issued: | |||
| Approved by: | Date: | Review Date: | |||
Reduced Testing Program
Introduction
This guidance document defines a science and risk-based approach for the evaluation and implementation of a reduced testing program for the release of starting materials,
intermediates, APIs, excipients and packaging components at a GMP user site upon receipt from vendors (manufacturer/supplier). The guidance also provides an example of the application of Quality Risk Management principles in the implementation of a reduced testing program.
There is limited value in testing starting materials, intermediates, APIs, excipients and packaging components previously released from vendors or other sites whose
operations are in a state of compliance with cGMPs and where the materials have reliably supported the production/supply of goods that meet the site’s in-house specifications.
The chief benefits of reduction of in-house testing of starting materials, intermediates, APIs, excipients and packaging components upon receipt are reduced lead times for
production and significant reduction in resource utilization (test reagents, equipment, analysts, reviewers, etc.). Therefore, each site should assess opportunities to eliminate
in-house release testing of raw materials and packaging components on a case by case basis. A review of organizational directives (constraints), in regards to sourcing of
materials, should be reviewed prior to initiation of any assessment.
Recommendations & Rationale for Recommendations
For reducing analytical testing, there are two (2) categories for consideration:
(1) Analytical tests performed only at a user site and
(2) Duplicate analytical testing performed by both user site and the vendor.
To address these categories, the following are the recommended approaches for reduced testing of each category:
(1) Eliminate unnecessary testing at the user site
(see section A)
(2) Accept an approved vendor’s test results in lieu of testing by the user site when both user and vendor perform testing (see section B)
For any approach, documentation of the review of data and the decision made, using Quality Risk Management principles must be prepared and approved using local or regional procedures prior to the elimination of any testing.
Special attention must be given to materials intended for use in sterile applications based on global regulatory requirements, as some countries/regions may not allow reduced testing of materials used in sterile products.
Consideration must also be given if the material is used in more than one product. In this case, the more conservative approach needs to be adopted unless the material is shown to be non-critical to the quality of all affected products.
- Elimination of Analytical Tests Performed Only at the User Site
(see Figure 1)
(1) Determine if the test is necessary from a scientific perspective
(e.g., is the data obtained required in order to provide confidence of a satisfactory
process and/or product(s).
- If the test is scientifically necessary, continue testing.
- If the test is not needed, determine if it is required in any regulatory
application.
(2) If the test is confirmed to be a regulatory requirement, evaluate if the
effort (resources, variation fees, etc.) required to make the regulatory
change results in an overall benefit.
iii. If a regulatory change will provide significant benefits, initiate internal and external change control procedures to eliminate the test with supporting documentation justifying the change.
- If there is no significant benefit, continue testing.
(3) If the test is a non-regulatory requirement and is considered not to be scientifically necessary, initiate change control procedures to eliminate the test, with supporting documentation justifying the change.
Figure 1: Regulatory and Non-regulatory Tests Elimination Process Flow
- Acceptance of a Vendor COA (refer to Figure 2)
(1) Materials from other sister sites
For materials manufactured, tested and released by other sister sites, testing at the receiving site may be reduced to visual inspection and ID testing from only one
(1) Container of each lot received (refer to Appendix A for scope and rationale) including those materials intended for marketing in the European Union (EU).
Note: Visual inspection includes labeling verification and confirmation that the material has not been tampered with.
(2) Materials from external suppliers, where both user site and vendor perform testing
- Verify that the vendor is an “Approved supplier”.
- Verify that the vendor has demonstrated an acceptable quality history based on a comparison of the supplier’s Certificate of Analysis (COA) results and site incoming test results.
- A minimum of 3 consecutive lots tested by user site must be used for evaluation.
- If the user site results met the specifications for all lots and nothing else was observed that calls into question the reliability of the vendor’s results, user site testing may be eliminated and materials received based on ID testing and vendor COA.
- If there was a confirmed failure of any lot tested by user site in the past year or the prior 3 lots fully tested (whichever is the higher number of lots), then the supplier cannot be deemed reliable. Verification that appropriate corrective actions have been taken by the vendor to address the failure must be performed. Testing by user site must be continued until at least 3 consecutive lots received meet user site specifications.
- No statistical comparison of user site results versus supplier results is necessary as long as the user site specification is the same as or wider than the vendor specification. If the user site specification is tighter or different than the vendor’s specification, a statistical comparison of user site results and vendor results will be necessary.
- If sufficient data has been generated regarding the material and vendor under review (as outlined above), it is not necessary to have information whether or not the vendor performs skip lot testing in order to make a decision on the supplier’s reliability.
- In rare instances, there may be regulatory commitments that indicate that user site will perform certain testing on incoming goods from vendors. If it is confirmed that a regulatory commitment was made:
- Evaluate if the effort (resources, variation fees, etc. required to make the regulatory change results in an overall
benefit
- If the regulatory change has a significant benefit, initiate internal change control procedures to eliminate the test with supporting documentation justifying the change.
iii. Once all requirements have been satisfied, reduce testing to COA review, ID testing and visual inspection unless additional testing is warranted for business purposes.
- One (1) lot of the material must undergo full testing annually.
- For materials used for products intended for the EU market, it may be acceptable to perform ID testing from only one (1) container of each material receipt, provided that the requirements of Annex 8 are met and the assessment properly documented. As outlined in Annex 8, the assessment for eligibility should cover both the material and its
vendor.
Figure 2: Acceptance of Certificate of Analysis Process Flow
Example of Quality Risk Management Process Application
(Note that other alternative approaches are equally acceptable) A Quality Risk Management approach is illustrated in this guidance to help evaluate the feasibility of reducing or eliminating incoming release testing of individual lots of starting materials, intermediates, APIs, excipients and packaging components. The approach identifies the different risk factors to consider when performing the evaluation and how to group potential risks into low, medium, or high categories.
For the purpose of this evaluation, three risk components, severity, probability and detection, will be examined for each material identified. From this evaluation, a list of potential materials (starting materials, intermediates, APIs, excipients and packaging components) for reduced testing will be developed. Each material will be considered a risk.
Through application of a simple tool coupled with requisite background knowledge, it is expected that this assessment will serve as a model for GMP sites to standardize the evaluation process for reduction of release testing of starting materials, intermediates, APIs, excipients and packaging components.
Organization of Information
The assessment will consider the following data with respect to a specific starting material, intermediate, API, excipient or packaging component:
- Criticality of the material to the overall performance/quality of the finished
product
- Regulatory commitments/expectations
- Supply history of a vendor for a specific material
- Quality History for a specific material
- Audit history of vendor
- Process monitoring/product release strategy
Risk Question
In this case, the desire for implementation of a reduced testing schedule or reduced testing program results in the following risk questions:
- “What are the analytical tests that can be eliminated or reduced for incoming
material receipts with the least impact to product quality or regulatory
compliance?”
- “What are the product quality and regulatory compliance risks associated with
accepting incoming materials using a vendor’s CoA?”
- “What are the starting materials, intermediates, APIs, excipients and packaging
components that can be transitioned to a reduced testing schedule with the least
impact to product quality?”
Risk Assessment Tool
Based on the data to be used for the assessment, an enhanced Risk Ranking and Filtering method (which considers detection in addition to severity and probability) is adopted to aid in determining those starting materials, intermediates, APIs, excipients and packaging components that can be transitioned to a reduced testing schedule without assumption of detrimental levels of risk (third risk question).
Risk Ranking and Filtering (RRF) typically focuses on two separate risk components, severity and probability, associated with each potential risk relevant to an issue. However, in this instance, the ability to detect (detection) nonconforming materials within the existing operation prior to release of the finished product (API, DP or Medical Devices) will also assessed.
Risk Assessment –Identification, analyses, and evaluation of potential risks
The following risk factors were identified as having an impact related to a reduction or elimination of release testing for starting materials, intermediates, APIs, excipients and
packaging components:
Impact on Product Quality/Performance – the criticality of the material to the overall quality and performance of the finished product. The failure of a material to conform to specifications could result in the finished product not achieving the desired therapeutic effect or meeting specifications throughout shelf life.
Regulatory Compliance -it is suggested that the expectations for the most stringent market served be used for the assessment when multiple markets are involved, since requirements vary from one market to another. Furthermore, efforts should be made to identify any specific commitments related to release testing that may have been made to a particular regulatory agency either through correspondence or within an approved filing.
Business Risks – any impact to the overall business may be assessed. Decisions on reducing or elimination of release testing for incoming materials may require
considerable resources to perform proper risk assessment or pose business risks, e.g. product rejection or missed production schedule due to investigations or failures obtained during finished goods testing when a required test was not performed, etc.
The stated risk factors related to a change in the release testing of starting materials, intermediates, APIs, excipients and packaging components must be considered during the risk assessment. As identified previously each potential risk (i.e. material) has an associated severity, probability and detection. Table 1 below represents the suggested scale (differentiations) for each of the three risk components to be used in the assessment to answer the third risk question:
-
- “What are the starting materials, intermediates, APIs, excipients and packaging components that can be transitioned to a reduced testing schedule with the least impact to product quality?”
To address the remaining risk questions, a variation to the descriptions of the suggested scale can be made. In this scenario, it is assumed that the acceptance of goods based on vendor COA is an accepted practice. As a result regulatory and business risks will not be evaluated. The severity will represent a measure of the material’s criticality to the overall performance
and quality of the finished product. Prior to making this determination it is suggested that relevant development reports, CMC sections and transfer reports be reviewed to assure
adequate understanding of the material’s role in the products formulation or packaging design. The probability will be determined as a combination of two factors: supply performance (numbers of lots received without issue) and the vendor’s compliance status as determined through periodic audits conducted by the site.
The detection will reflect the opportunities presented in the production and release testing processes to assess the performance of the material in the associated finished product(s). Opportunities for detection may include in-process tests or checks, either manual or automated, throughout the defined production process or release testing, for each process step or finished product. In some instances a mode of detection may be the actual design or configuration of process equipment. Equipment that is set-up to exacting tolerances and is subject to jamming upon attempted processing of non-conforming goods may be viewed as an opportunity for detection.
The example illustrates the use of “numbers” rather than “words” to rank the differentiations between risk components. It is important to note that each number
signifies a level of risk that could be alternatively expressed in words, e.g. Critical, Major, Minor etc. as shown in highlighted sections of Table 1. It should be noted that
scoring of risks during the assessment is based on the worst applicable condition associated with a given risk.
Table 1: Probability, Severity and Detection Ranking Scales
The tool is applied to the risks identified and a Risk Score is calculated using the values assigned for severity, probability and detection.
Probability x Severity x Detection = Risk Score
Risk Acceptance
After the Risk Score has been calculated for the individual potential risks it must be assessed versus an evaluation matrix to determine the acceptability of the existing risk or
conversely, identify the need for reduction of the risk through implementation of controls, where possible. The evaluation matrix is to be devised based on a site’s willingness to
accept different levels of risk (determined prior to conducting ranking of the various risks). Examples of the risk score evaluation matrices are shown in Tables 2a and 2b.
Table 2a exhibits preliminary scoring based only on the two components of risk: severity and probability. The interpretation of the risk associated with a particular material should
be based on Table 2b which incorporates the scoring indicated for all three risk components: severity, probability and detection.
For this example, an acceptance threshold value of 30 was selected. This means that any material evaluated that receives a score greater than 30 represents a high risk to
implement a reduced testing schedule based on the current conditions. This value was derived based on the willingness to accept reduced testing for API, critical excipients and
primary packaging and labeling resulting in a severity value of 3. A value of 5 was deemed the minimal acceptable level of probability, meaning we would expect that a
vendor would have at least a Conditionally Acceptable status and have demonstrated acceptable performance through the supply of at least three consecutive conforming lots
of the material. When viewing detection it was determined that having at least one IPC or finished product test to assess the material’s performance in the finished product was
desired, rendering a detection value of 2. When the values for each of the three risk components were multiplied together (3 x 5 x 2 = 30), a total risk score of 30 was
obtained. Additionally, any total risk scores derived using a probability value of 10 would represent an unacceptable level of risk. Vendors are given a probability score of
10 when they have been shown to be unacceptable through audit and/or supply performance or have undertaken recent change in their production process rendering past performance data meaningless.
Table 2a: Preliminary Risk Score Evaluation Matrix
Table 2b: Total Risk Score Evaluation Matrix
Risk Control
For those risks that are deemed to exceed the site’s risk acceptance threshold mitigation steps must be taken before proceeding forward with a change in release testing process.
For this example, most mitigating actions will target the probability and detection components of risk. A site may choose to work with a vendor to improve the compliance
status of the facility supplying the material or aid in improving their process to improve the quality of the material received; both actions target the probability associated with a
material. For mitigating a high detection score a site may implement in-process checks or tests to monitor performance of the material. Only when all risks are reduced to meet
the site’s pre-defined acceptance threshold should the process proceed forward. This should be confirmed via re-application of the tool for risks that were the subject of
mitigation efforts.
Risk Review
When all risks are judged to comply with the pre-establish risk acceptance level, the document will sent to the Quality Team for review and approval. The documentation
package should contain all documented aspects of the Quality Risk Management process. Implementation of the proposed change in frequency cannot proceed until all approvals
are obtained. The risk assessment process should be repeated any time a change is introduced that impacts the practice, e.g. change in vendor’s audit status or change in
vendor’s manufacturing site, process, grade of material sourced, etc.
As stated previously, when a reduced testing schedule is implemented the site should minimally perform full testing on at least one lot of the material each year. Additionally,
laboratory investigations and quality assurance reports should be monitored for any instances related to the material. Upon observation of quality issues traceable to a
specific material a reassessment should be performed to identify the risk associated with continuing a reduced testing schedule for the material.
QRM methodology can also be used in reference to reduced testing to aid in prioritization of assessment based on factors such as volume of lots tested per year, time/resources
required to complete full testing, complexity of testing, etc.
APPENDIX A
ID Testing From One (1) Container of a site Manufactured Starting Material/Component or Finished Goods
PURPOSE
The purpose of this paper is to document the rationale for performing incoming ID testing using one (1) container of a site-manufactured lot.
SCOPE
This document can be applied to site-manufactured intermediates, APIs, bulk and/or finished drug products. This document does not apply to materials sourced and manufactured by other companies regardless of their approval status.
RATIONALE
Current Good Manufacturing Practices (cGMPs) require that the identity of components1 (starting materials2) such as intermediates, active pharmaceutical ingredients (APIs), bulk and finished drug products must be verified prior to use.
21 CFR 211.84(d)(1) states that “At least one test shall be conducted to verify the identity of each component of a drug product. Specific identity tests, if they exist, shall be used.” Per 21 CFR 211.84(b), the regulations allow representative sampling of each shipment of each lot for testing or examination.
Chapter 5, section 5.30 of the European GMPs requires the availability of “…procedures or measures to assure the identity of each container of starting material.”
Per Annex 8 of the EU GMPs, it is permissible to sample only a proportion of the containers “…where a validated procedure has been established to ensure no single container of starting material has been incorrectly labeled.” Annex 8 further stipulates that: “Under such a system, it is possible that a validated procedure exempting identity testing of each incoming container of starting material could be accepted for:
1 From 21 CFR 210.3(b)(3) as “any ingredient intended for use in the manufacture of a drug product, including those that may not appear in such drug product.” 2 From EU GMP Glossary as “any substance used in the production of a medicinal product, but excluding packaging materials.”
– starting materials coming from a single product manufacturer or plant;
– starting materials coming directly from a manufacturer or in the
manufacturer’s sealed container where there is history of reliability and
regular audits of the manufacturer’s Quality Assurance system are conducted
by the purchaser (the manufacturer of the medicinal product) or by an officially accredited body.”… 3
Based on the FDA and EU regulation sections cited above, only one (1) container of each shipment of a component (starting material), API, bulk and finished goods lot received by
a GMP site, manufactured and shipped from other sister plants/sites, will be used for identity testing (ID).