Introduction
This guidance describes the documentation needed to support the use of Continuous Quality Verification (CQV) to demonstrate that a drug product (DP) or active pharmaceutical ingredient (API) process is in a validated state. It also describes some similarities and differences between CQV and traditional process validation using three discrete lots.
Process validation is used to provide assurance that the processes used to manufacture pharmaceutical products result in products which possess the required critical quality attributes (e.g. strength, identity, purity, safety, and efficacy).
Continuous Quality Verification (CQV) is an alternative approach to process validation. One of the primary differences for CQV compared to a conventional discrete, 3-batch process validation approach is that the process is continually monitored, evaluated, and adjusted (when necessary) to achieve defined Critical Quality Attributes (CQAs) using validated in-process measurements, tests, controls, and process end points.
CQV is applicable to all types of products and processes (e.g. Biopharmaceuticals, API, and drug product) and may be used with new, legacy, batch and continuous processes.
The principles may be applied during development of a new process or product, or for the improvement and/or redesign of an existing process. CQV may be applied to an entire process, or to defined critical manufacturing steps / unit operations.
Using CQV, processes are continuously verified as being capable of providing the desired product quality rather than the reliance on data generated from a few production lots.
Recommendations and Rationale
The CQV approach to process validation may be applied to both new and legacy processes when the necessary information is available. It requires a good understanding of the process and a process control strategy that ensures repeatable and robust performance of the process.
Differences between processes implementing CQV and those following a conventional approach:
The elements of CQV:
There are four main elements needed for CQV. Some aspects of these are also necessary for conventional process validation.
As summarized in Figure 1 below; each of these elements requires some form of documentation or records. Collectively these documents will provide the necessary evidence to show that the process operates in a validated state, and the ongoing monitoring, control and analysis provides assurance that the process continues to operate in a validated state.
1. Process Understanding:
Process design documentation is a pre-requisite for CQV and should include the following:
a) Documented summary of scientific understanding of Product and Process(es)
During process development, the proposed process design is investigated and characterized by experiments, process modelling, etc. Critical Quality Attributes (CQAs) for the product must be defined and documented. Potential Critical Process Parameters (CPPs) to be controlled and monitored in order to achieve CQAs must also be identified whether using a conventional or a CQV approach The rationale for CPPs and non-critical parameters should be documented in the form of a risk assessment, including the probable adverse consequences when the CPP ranges are exceeded. (Identification of CPPs is described in the guidance for, “Evaluation and Classification of Process Parameters as Critical” and other associated guidance.)
The risk assessment to define and justify the final CPPs and CQAs may be an iterative process as the understanding of the process increases.
The summary of the process understanding may also include enhanced understanding of material attributes, processing options and process parameters and be documented as Design Space (ref. ICH Q8).
Data supporting process understanding may be obtained from development and scale-up studies, experience from similar products and processes, or may include data from commercial scale manufacturing, depending on the stage of the product lifecycle.
The responsibility for providing the data will depend on the stage of the product lifecycle; for example, Research and Development (R&D) should provide documentation (for example, a Process Knowledge Report or Process Understanding Plan) for a new process, whereas for a legacy process, commercial batch data and any experimental data may be provided by the site Technical Services or Production Support groups.
b) Documented Review of revised Risk Assessment review
A review of the Risk Assessment defining the final CPPs and CQAs should be conducted once the specific site and equipment where the process will be performed are identified.
In addition to accepting the risk assessment justifying the CPPs and CQAs, the review should include consideration of:
The level of process knowledge available to support the proposed commercial scale (e.g. scalability studies or data at full scale)
The impact of equipment capabilities on the CPPs that have been identified (e.g. comparison of expected normal operating ranges with proven acceptable ranges)
This revised risk assessment is recommended for any process, whether using CQV or a conventional process validation approach. This revised risk assessment is site-oriented, and therefore, where practical, personnel from the assigned manufacturing site(s) (for example, site technical services, site production support) should be involved in the preparation. It should be finalized before the approval of the Process Control Strategy as it is a key input to the strategy.
c) Process Capability Studies
It is recommended that CQV is only applied to processes that have been demonstrated to be both capable and stable. The data for this evaluation may also be used for the Preliminary Performance Evaluation (see below). Right-First-Time8 (RFT) tools should be used to evaluate the process; the type of tool used may depend on the amount of data available.
For legacy processes where implementation of CQV is being considered, the Right First Time initiative on Product Based Process Capability may provide an indication of the process performance and therefore its suitability for CQV implementation.
2. Continuous Quality Monitoring & Control.
a) Process Control Strategy
This document defines the process control strategy proposed for routine manufacture, describing the scheme for monitoring, measuring, analysing and adjusting (when necessary) the critical aspects of manufacturing steps / unit operations, and how this scheme will ensure control of product quality. Real-or near-time adjustment of the process, when necessary, is a key element of CQV.
The control strategy defines the planned set of controls, derived from current product and process understanding that assures process performance and product quality during routine manufacture . It should describe the input material controls, process controls and monitoring, and finished product tests, as appropriate, that are proposed and justified in order to ensure product quality.
It is recommended that a more detailed, site-specific process control strategy is prepared by site personnel (e.g. technical services, quality teams). This can be used to replace the conventional process validation protocol and, if so, must be approved prior to manufacture of the commercial batches.
The site-specific process control strategy should include or reference:
The steps/unit operations included in the scope of the strategy document
The critical quality attributes and/or critical process parameters that need to be monitored and controlled, and references to the most current risk assessment.
How the CQAs and CPPs will be monitored, evaluated and controlled, for example:
- Drug substance and drug product materials and components
- Facility and equipment operating conditions
- In-process controls
- Finished product specifications
The tasks required to assure the process is valid and suitable for commercialization include:
- Determine the number of lots to be included in the preliminary process performance evaluation. It is recommended that an explanation be provided for the number of lots that will be used to demonstrate the performance of the process.
- Document a description of any one-time studies (e.g. content uniformity, homogeneity, equivalency comparison) and their acceptance criteria.
Continuous analysis of the process may be achieved through a number of methods. “Continuous” is not intended to suggest that uninterrupted, on-line analysis is required; the time period between sampling/analysis should consider both the rate of change of the process and how quickly an adjustment to the process trajectory can be made.
Process measurement techniques may include conventional data collection plans, process control charts, production record data and Process Analytical Technology (PAT) systems.
These may allow for the collection and further analysis of real-or near-time data, for example, in-process and/or final product CQAs, process end-points and CPPs.
b) Pre-determined acceptance criteria and specifications
There must be documentation that defines the pre-determined, approved acceptance criteria for all critical parameters, controls and attributes that provide assurance that the identified CQAs will be consistently met during commercial manufacturing. It is often included as part of the control strategy. Acceptance criteria may also be documented in product / (in)process specifications.
c) Batch Record Data
For both the conventional process validation and the CQV approach, the batch record is based on the approved process control strategy, together with the relevant acceptance criteria and specifications. All quality monitoring and control data required to demonstrate that the process is performing as required should be included or referenced in the batch record.
3. Process Analysis
a) Preliminary Process Performance Evaluation
At the start of commercial manufacturing, before distribution of the finished drug product, a documented, process performance evaluation is required that confirms that the process design achieves the desired results. This evaluation is necessary, whether using a CQV or a baseline approach. It is typically prepared by site technical services and/or operations personnel, with approval by Quality. It also corresponds to the FDA “conformance batch” phase5. The batches are manufactured under normal processing conditions.
The Preliminary Process Performance Evaluation should include or reference conclusions from any one-time studies verifying that the process design meets expectations such as content uniformity, homogeneity and/or mixing studies, as appropriate. These studies should be evaluated against pre-approved acceptance criteria.
The Preliminary Process Performance Evaluation may also include a review of non-routine data from the process in order to provide further verification of the process performance and confirmation that the CPPs identified are appropriate.
The Preliminary Process Performance Evaluation should be described in a pre-approved protocol or plan that defines the expectations and acceptance criteria. The plan or protocol may be a separate document or it may be combined with the process control strategy.
Once the defined data have been collected for the required number of batches, a report should be provided that includes the following:
A review of the manufacturing data for CPPs and CQAs against the acceptance criteria
An evaluation of the process performance, including analysis of process capability. It is recommended that statistical process capability analysis, if used, be modelled on practices recommended by the Right First Time program and should include assessment of attributes that are critical to product quality.
A conclusion whether the process is considered validated and recommendations for any modifications to the process understanding (e.g., CPPs, Design Space) or control strategy based on the increased process understanding acquired during the performance evaluation.
A recommendation on the appropriate frequency for routine process performance evaluation, the data to be reviewed and how the data will be analysed.
b)Ongoing process monitoring and analysis
Process monitoring documentation typically includes manufacturing records and other process measurement data, as defined by the process control strategy.
When using a conventional process validation approach, individual batch release is based on confirming that the fixed process established during process validation has been repeated and that relevant in-process and end-product test specifications have been met.
For a process using CQV, individual batch Acceptance and Releaseis based on evaluation of the process data and process performance at a pre-defined stage of the process. The implementation of CQV may provide a level of confidence that each batch conforms to established quality attributes to enable the real time release of the final product, or to justify reduced end-product testing.
The acceptance criteria for release should include confirmation of the validity of the manufacturing process for the specific batch. The validity of the process is based on confirming that the process remained within the acceptance criteria defined in the control strategy and that this level of process control delivered the required product quality attributes.
In addition to the review of process performance that takes place as part of the release of an individual batch, there should also be ongoing monitoring and evaluation of the process over multiple, commercial batches. Ongoing monitoring of process performance can be achieved through a periodic documented evaluation. This periodic evaluation may be conducted as part of the Annual Product Review, or may be more frequent to verify the continued performance capability of the process.
4. Continuous Process Improvement
All processes should be reviewed to identify opportunities for continuous improvement, such as product quality improvements, process improvements, variability reduction, innovations, and pharmaceutical quality system enhancements, thereby increasing the ability to consistently fulfil quality needs . The fact that CQV requires continuous verification of the quality of the process ensures that relevant process performance data are routinely collected.
To support continuous improvement, the results of any process evaluation should be fed back into the process design. This increases process understanding and supports optimization of the process throughout the product lifecycle. The application of statistical tools to monitor the process capability is recommended when sufficient data are available. It is recommended that the use of appropriate RFT tools be considered to support this ongoing process performance evaluation and to highlight potential areas for improvement.
a) Change Management Documentation
Modifications to the documented process understanding (for example, Process Understanding Plan, Design Space, CPP/CQA risk assessment) and the process control strategy should be managed and documented using an appropriate change management system. It is recommended that the existing site change management system is used, updating the procedure if necessary – a separate change management system is not required or recommended.
Whether using a conventional process validation approach or CQV, any change should be assessed to determine the impact on the validated process and identify any associated validation activities.
It is recommended that unplanned changes to the process, or deviations, should also be investigated and documented according to existing site procedures. The investigation should include consideration of the impact of the deviation on the process control strategy, in addition to the product quality.