A Case for Stage 3 Continued Process Verification

Reducing process related failures and quality costs; Continued Process Verification can improve control and lower failure rates

By Daniel Alsmeyer and Ajay Pazhayattil, Apotex Inc.

2 of 3 1 | 2 | 3 View on one page

Stage 3A is designed to allow close monitoring of parameters and quality attributes and to detect any undesirable process variability trends observed post launch, thus providing an opportunity to make recommendations and finalize the routine Continued Process Verification (Stage 3B) control limits. Stage 3A is used to perform additional sampling and testing (monitoring), if required, based on the Stage 2 assessment. Stage 3A is performed under a pre-approved protocol. Stage 3A Continued Process Verification Protocol uses statistical measures to determine process performance. Process performance (Ppk) values are computed based on a predetermined number of consecutive commercial batches. Ppk values and Statistical Process Control (SPC) charts are developed for in-process critical quality attributes (CQA) and critical process parameters (CPP) where applicable. Similarly, process performance capabilities and probability of Dissolution Acceptance (Pa) are determined for the Finished Product critical quality attributes. A Stage 3A Report includes a discussion around any excessive intra-batch variability.

Accepted statistical guidelines indicate processes with Ppk more than 1.0 are within statistical control. Appropriate actions (changes, continuous improvement activities, etc.) are discussed in the report if Ppk values are less than 1.0. The Stage 3A statistical analysis provides a science and risk-based assessment of the product meeting the critical quality attributes in the future.

Table 2 shows how compaction parameters might be adjusted post Stage 3A to consistently meet the established quality attributes. The continued batch to batch verification at Stage 3B will further highlight any undesired trends.

A decision to re-validate (Stage 1 and Stage 2) the commercialized manufacturing process starts with the change control process. Supporting data for the change and a technical risk assessment document the process re-validation strategy. Examples of process changes where re-validation may be recommended are:

• Investigation/failure related change recommendations
• Formulation and/or process changes such as scale up/scale down
• Continuous process improvement recommendations
• Material and/or equipment changes
• Regulatory notifications, etc.

Batches are not released for commercial sale until Stage 2 is successfully completed for the change (Refer to Figure 3 - Stage 1, 2, 3, Changes).

Continued Process Verification (Stage 3B) trend reports provide a statistically relevant and current capability score. This information is used for deciding a validation strategy. For example, a lower dissolution probability (low probability of meeting the USP Stage 1 dissolution criterion) may warrant more batches and additional sampling at Stage 2 re-validation to ensure the change did not adversely impact dissolution results.

Figure 4 displays a statistical model developed to determine the minimum number of batches required for Stage 2 re-validation on the basis of the products current capability score. Stage 3B is used to establish a scientifically justified sampling and testing plan while performing the risk assessment and drafting of the Stage 2 re-validation protocol. A review of Stage 3B trend reports aid in gaining knowledge and performance insights of similar process/products. This data is relevant in continuous improvement and determining optimal processing parameters through product development and scale-up activities.

FDA Guidance Process Validation: General Principles and Practices (P. 11) recommend use of credible experience with sufficiently similar products and processes. The cumulative data from all relevant studies (e.g., designed experiments; laboratory, pilot and commercial batches) are used to establish Stage 2 requirements. Bracketing of multiple strengths may be possible with an assessment that demonstrates low risk due to the change based on the Stage 3B trend. A possible candidate for such an approach is wherein an API change has a minor impact to the process and there is very solid Stage 3B trending. A bracketing approach to Stage 2 re-validation, such as three batches of the high and low strengths and one each of the middle strengths, may also be employed.

Another measure of interest, especially during development of a re-validation strategy, is the Stage 3B capability score for routine blend and dosage uniformity results. A low score may indicate a potential segregation concern. The Stage 2 re-validation sampling and testing plan must be developed in a manner that, though low, mitigates the identified risk. A risk-based (stratified beginning and end) sampling plan and an example outcome is provided in Figure 5.