The validation data review is the process of critically analyzing the data generated during validation or qualification studies with the objective of checking whether the study was conducted according to the protocol, whether the data obtained is valid, whether the acceptance criteria were realized and whether the conclusions are based on scientific understanding. In addition to simply summarizing the review findings, Quality Assurance (QA) should verify that the data confirms the findings.
In the pharmaceutical production world, validation data review can include process validation, cleaning validation, equipment qualification, utilities qualification, analytical testing validation or validation of computerized systems. Regardless of their differences in technicalities, the question at the core of each review process remains unchanged - is the documented evidence proving that the process/system meets its specifications?
Check that the study carried out is in accordance with the approved protocol, in terms of its revision number, equipment or system title, particulars on the product used if any, as well as planned work. Confirm that the protocol was approved before the study was executed according to the site's policies.
Extra caution should be applied when actual practice is different from what has been planned. For instance, sample site could not be reached, experiment could have been conducted outside of desired time frame or some conditions could have been changed while taking measurements. The deviations should be reported, evaluated and addressed according to appropriate deviation and change regulation plan.
QA must not accept the reports solely based on the final result being within acceptable limits. Once it has been established that the study has not been carried out in accordance with the permission granted, it is crucial to evaluate the effect of such deviation on final results validity.
The checks must provide traceability of the report and protocol, laboratory logs, machine records, electronic files, sampling logs and other relevant proofs. The results must possess identifiable sources that would include the sample number, run or batch, machine, testing day, analyst or operator and any other identifiers required by the guideline used.
In one instance, a process validation report may contain a summary of the blend uniformity results from different sampling locations. It’s the responsibility of QA to ensure that the values reported match the samples, that all the required locations have been sampled and that the analytical results justify the summary made.
Pay attention to the missing pages, duplication of the results, discordant identifiers, corrections not explained, unfilled worksheets, or difference in the databases and final report. In case the result is transcribed to the electronic spreadsheet, it shall be checked if both the transcription and the calculations are correct.
The analysis should be performed on the risk basis, but it cannot become just an outer verification of the easily seen successful results while the important data may stay unnoticed.
It is necessary to evaluate individual results in addition to the total pattern. The average score might give a wrong impression because some isolated result might require a follow-up. The same applies to results that are in compliance with the acceptance limits but show a significant trend or variation that needs scientific consultation.
The type of review to be performed will depend on the study at hand:
The ALCOA+ principles should be applied. The data should be attributable, legible, contemporaneous, original, accurate and there should be relevance related to completeness, consistency, endurance and availability.
The printed summary for electronic systems doesn’t provide sufficient evidence alone. Depending on the system, study and procedures used, QA may need to check the native electronic records, metadata, audit trails, history of processing and related records of changes or accesses.
For instance, if the chromatographic data were reanalyzed after the first result, the reviewer should determine what changed, who acted, whether justification was made and what effect the changes had on the result. The original data and history of processing should be available to the reviewer in keeping with the rules being applied.
Among other red flags are missing files without explanation, running experiments again without any justification, altered calculations or discrepancies between digital and paper record. All these should be examined by professionals according to data integrity procedures at the site.
For every situation in relation to the validation process, check if the documentation contains a report about the case, listing what happened, what activity or data was being impacted, what influence it had and what was the investigation carried out about this case. Make sure the CAPA is pertinent to the problem and that the responsibilities of different people regarding the situation are stated properly.
The process of repeat testing needs careful evaluation. A successive successful result does not mean the previous failure is irrelevant. The reasoning behind the repeat testing, the status of the former results and the final conclusion need to be justified based on science and regulatory techniques.
If any considerable concern remains unresolved, QA office needs to clarify if its report states that objectives of the investigation have been accomplished. In case of lack of proofs, further investigations, verifications, or any other warranted actions are needed.
The report might declare successful qualification of the equipment, while not being able to demonstrate that the manufacturing process always results in the product being permissible. Similarly, passing finished product tests does not mean that any goal of the validation process has been reached.
Prior to the approval ensure that:
- All expected tests and records are mentioned
- Any deviances and unique data are presented
- The conclusion supports the reality and any restrictions
- The pending issues are stated and controlled
- The document is complete, versioned and signed
An inspector might choose a datum from the final report and ask QA to back-track it up to the primary source, analyze a strange tendency, or give explanations on how a failed test influenced the conclusion. The reply must be based on the records, rather than confirmations in words.
The standard procedures for validation must clearly outline who conducts the examination, how comments are going to be dealt with, who is going to escalate discrepancies, who has the power to sign the document and make sure that the record keeping procedure has been defined. Technical experts have to evaluate appropriate evidence, while the QA is responsible for making sure that the documentation and its final decision is acceptable.
Begin With the Approved Protocol and Study Scope
Prior to analyzing the specific outcomes, QA needs to understand what the study was intended to show. The approved protocol stipulates all the parameters, tasks, testing conditions and documentation needed.Check that the study carried out is in accordance with the approved protocol, in terms of its revision number, equipment or system title, particulars on the product used if any, as well as planned work. Confirm that the protocol was approved before the study was executed according to the site's policies.
Extra caution should be applied when actual practice is different from what has been planned. For instance, sample site could not be reached, experiment could have been conducted outside of desired time frame or some conditions could have been changed while taking measurements. The deviations should be reported, evaluated and addressed according to appropriate deviation and change regulation plan.
QA must not accept the reports solely based on the final result being within acceptable limits. Once it has been established that the study has not been carried out in accordance with the permission granted, it is crucial to evaluate the effect of such deviation on final results validity.
Trace Reported Results Back to Original Data
The validation report typically contains chosen results, analyses, tables and conclusions. The quality assurance team must check that summaries truly reflect the primary records.The checks must provide traceability of the report and protocol, laboratory logs, machine records, electronic files, sampling logs and other relevant proofs. The results must possess identifiable sources that would include the sample number, run or batch, machine, testing day, analyst or operator and any other identifiers required by the guideline used.
In one instance, a process validation report may contain a summary of the blend uniformity results from different sampling locations. It’s the responsibility of QA to ensure that the values reported match the samples, that all the required locations have been sampled and that the analytical results justify the summary made.
Pay attention to the missing pages, duplication of the results, discordant identifiers, corrections not explained, unfilled worksheets, or difference in the databases and final report. In case the result is transcribed to the electronic spreadsheet, it shall be checked if both the transcription and the calculations are correct.
The analysis should be performed on the risk basis, but it cannot become just an outer verification of the easily seen successful results while the important data may stay unnoticed.
Review Calculations, Acceptance Criteria and Trends
The QA group has to check if the results have actually been evaluated using the defined acceptance standards from the protocol. In case of calculations, the reviewer should verify the formula, units, rounding rules and raw values, particularly in the cases when calculations impact the evaluation of compliance.It is necessary to evaluate individual results in addition to the total pattern. The average score might give a wrong impression because some isolated result might require a follow-up. The same applies to results that are in compliance with the acceptance limits but show a significant trend or variation that needs scientific consultation.
The type of review to be performed will depend on the study at hand:
- For process validation, the review would check the critical process parameters, critical quality attributes, in-process results, testing of the finished product and consistency of the results across batches.
- In cleaning validation, the review would check sampling points, considerations regarding the recovery of the swab/rinse, suitability of the analysis, residue calculation results and compliance with the specified acceptance limits.
- In the case of qualification of the equipment and utilities, it is important to verify that the tests done, operating ranges, alarm systems, interlocks and performance checks were done as per the criteria set beforehand.
- For analytical method validation, the review would check the relevant characteristics of validation, calculations, chromatogram, equipment ability and the report on unpredicted results.
Check Data Integrity and Electronic Records
The importance of data integrity cannot be overstated when discussing the validation process. It falls on quality assurance to ensure that records’ data is reliable, it should check if the data is complete, consistent, reliable and traceable.The ALCOA+ principles should be applied. The data should be attributable, legible, contemporaneous, original, accurate and there should be relevance related to completeness, consistency, endurance and availability.
The printed summary for electronic systems doesn’t provide sufficient evidence alone. Depending on the system, study and procedures used, QA may need to check the native electronic records, metadata, audit trails, history of processing and related records of changes or accesses.
For instance, if the chromatographic data were reanalyzed after the first result, the reviewer should determine what changed, who acted, whether justification was made and what effect the changes had on the result. The original data and history of processing should be available to the reviewer in keeping with the rules being applied.
Among other red flags are missing files without explanation, running experiments again without any justification, altered calculations or discrepancies between digital and paper record. All these should be examined by professionals according to data integrity procedures at the site.
Evaluate Deviations, Failed Tests and Unexpected Results
You should never evaluate a validation study by just checking the results for passing grades. Aspects that must be taken into consideration before QA can conclude that the research was valid include deviations from the validation study objectives, failures to achieve positive outcome, rare observations and alterations in the validation procedure.For every situation in relation to the validation process, check if the documentation contains a report about the case, listing what happened, what activity or data was being impacted, what influence it had and what was the investigation carried out about this case. Make sure the CAPA is pertinent to the problem and that the responsibilities of different people regarding the situation are stated properly.
The process of repeat testing needs careful evaluation. A successive successful result does not mean the previous failure is irrelevant. The reasoning behind the repeat testing, the status of the former results and the final conclusion need to be justified based on science and regulatory techniques.
If any considerable concern remains unresolved, QA office needs to clarify if its report states that objectives of the investigation have been accomplished. In case of lack of proofs, further investigations, verifications, or any other warranted actions are needed.
Confirm that the Conclusion is Supported by Evidence
The final review must address the objectives, outcomes, divergences and acceptance standards of the investigatory process. QA has to ask themselves whether the report really proves its point, rather than claiming much more than the evidence can demonstrate.The report might declare successful qualification of the equipment, while not being able to demonstrate that the manufacturing process always results in the product being permissible. Similarly, passing finished product tests does not mean that any goal of the validation process has been reached.
Prior to the approval ensure that:
- All expected tests and records are mentioned
- Any deviances and unique data are presented
- The conclusion supports the reality and any restrictions
- The pending issues are stated and controlled
- The document is complete, versioned and signed
An inspector might choose a datum from the final report and ask QA to back-track it up to the primary source, analyze a strange tendency, or give explanations on how a failed test influenced the conclusion. The reply must be based on the records, rather than confirmations in words.
Make the Review Consistent and Defensible
A well-organized checklist for reviewing can assist the QA department in maintaining uniformity in applying standards in diverse validation practices. The checklist needs to fit the research rather than replace professional judgment. The review depth must depend upon the specifics of the process, risks associated with data, complexities of the system and the risk of consequences for product quality.The standard procedures for validation must clearly outline who conducts the examination, how comments are going to be dealt with, who is going to escalate discrepancies, who has the power to sign the document and make sure that the record keeping procedure has been defined. Technical experts have to evaluate appropriate evidence, while the QA is responsible for making sure that the documentation and its final decision is acceptable.
Get ready to use editable Validation Protocols in MS-Word FormatView List
No comments:
Post a Comment
Please don't spam. Comments having links would not be published.