The journey from an initial idea through to approval is challenging for any new medicine. Once a promising clinical lead has been chosen, the next step is to make GMP-quality material in larger quantities, using processes that are both robust and reproducible. While this can be difficult for a small molecule drug, it can be even more challenging for a biological drug because of the complex nature of the recombinant protein. It is, therefore, essential that the manufacturing process reliably makes batches that meet product specifications, every time.

Process validation is an important step for potential therapies that have had early success in the clinic and are progressing toward commercial manufacturing. The process validation strategy should be informed by validation guidelines laid down by regulatory bodies such as the U.S. Food and Drug Administration (FDA) and the International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use (ICH). For example, the 2011 FDA validation guideline¹ frames the three stages of validation as a continuum. This article will present a template for a repeatable late-phase development approach that leverages prior knowledge to speed timelines and enable the best chance of success.

At the beginning of Stage 1, when the molecule moves into development, a target product profile (TPP) has been defined that stipulates the attributes of the molecule that will make it an acceptable drug, considering both its safety and its efficacy. During Stage 1, a process is designed and defined that will allow clinical manufacturing of batches that are sufficiently large to supply Phase 1 and Phase 2 trials. This process should be carefully characterized to provide an insight into how the process (parameters and inputs) links to the chemical and physical attributes of the molecule that is being made.

In Stage 2, the growing knowledge about both process and product are elaborated upon, with the final stages of manufacturing equipment and facility qualification and process performance qualification (PPQ) being carried out. This will show whether the process is sufficiently robust to make the drug molecule consistently, and the data and control strategy will be part of the biological license application (BLA) filing for the drug. This is also the point at which the commercial manufacturing and life-cycle of the product should be considered in detail in preparation for commercial manufacturing, which occurs in Stage 3.

Regulatory strategy for process validation
ICH Q8² is a pivotal document, being one of the first to lay out the requirements for Quality by Design (QbD). This is a process which links what is known about the molecule, both in terms of its structure and its activity, creating the TPP, which leads to the definition of the product’s critical quality attributes (CQAs). These are the attributes of the molecule that can be measured and tested and are linked to patient safety and efficacy. It also contains risk assessments that include raw material attributes. This control strategy should cover everything that is added to the biomanufacturing process, including media and other components of the cell culture, materials required for the drug’s purification, and any excipients that are required for the drug product formulation.

The TPP is used to inform the late-phase design space, and the information and knowledge about the process is considered from the molecule’s standpoint to identify which aspects are critical, and how they might be controlled to ensure compliance and reproducibility. This control strategy, which is a crucial part of the regulatory filing, expresses how product specifications will be controlled, and also forms a part of the product’s life-cycle, offering an opportunity for constant process verification and continual improvement.

Previous experience with similar molecules can, and should, be used to inform these decisions. Whether the molecule in question is a monoclonal antibody, a fusion protein, or some other form of molecule produced by DNA expression, there is already a wealth of experience in testing and establishing stability. Initial results from in vitro and in vivo studies on the molecule, and early clinical safety and efficacy data, can also be invaluable here. This can all be linked to the product attributes, with a criticality assessment. All this information will inform the thinking about which critical quality attributes need to be controlled, how manufacturing will ensure that there is sufficient control, and how it will be monitored.

Critical quality attributes
A CQA is anything that can impact a drug’s safety, efficacy, and quality. ICH Q8 defines it as “a physical chemical, biological, or microbiological property or characteristic that should be within an appropriate limit, range or distribution to ensure the desired product quality.” All identified CQAs are linked to critical process parameters, based on the risk assessment that is carried out, and a final list of CQAs is generated to map the control strategy. A list of typical CQAs for a monoclonal antibody is shown in Table 1.

If the molecule is not yet well characterized in the clinic or by biochemical and other testing, an additional risk assessment should be carried out to identify any new CQAs and critical process parameters that should be included. Either way, process characterization and design of experiments (DoE) will be applied.

Determining the CQAs should be one of the first tasks in a process characterization. It is important to remember that not all molecules – even if they are the same type of molecule, such as a monoclonal antibody – will have the same CQAs. It is, therefore, essential to assess all the knowledge that has been gained about the molecule from early development work to ensure all the CQAs have been correctly identified.

Although it is possible to make some assumptions with regard to typical CQAs, they are very specific for the individual molecule and every CQA should be associated with critical process parameters to characterize the process and generate a control strategy. This control strategy is a regulatory requirement, and its main objective is to demonstrate that there is a process that will achieve the required quality for the product.

Based on a criticality risk assessment, a testing plan that is based on the criticality of product attributes should be established. Each CQA should have at least one analytical method, batch record instruction or in process control (IPC), and frequency of monitoring associated with it. Analytical control measurements can be made in-process, for release, after stability testing, and as part of the characterization process. Some CQAs will need to be checked at all of these four stages, such as the absence of aggregates. Others will only be required at certain points; Immunoglobulin G (IgG) concentration, for example, will need to be measured in-process and at release, and protein identification via peptide mapping will only be required as part of the characterization process.

Some critical process parameters are of a higher risk than others, and at the outset of a process, thoughtful consideration must be given to what represents “acceptable risk” and what does not. There are process parameters that might not impact any of the CQAs but may still affect process performance and will need to be evaluated. Scores from the risk assessment will be used to decide exactly which parameters will be evaluated during the process characterization experiment. Since all the experiments will be on a smaller scale than the final process, a scaled-down model for the process characterization work will be required.

Different types of experiments can be performed during process characterization, including one-factor-at-a-time (OFAT), multi-factor experiments and DoEs. Depending on the objective and knowledge of the process, there are multiple types of DoEs that can be performed and full factorial and fractional factorial are the screening designs that are traditionally used. Definitive screening design is a relatively new approach, which targets screening with some potential to identify optimization opportunities.  Ultimately, a risk matrix is defined for every operation, and many different studies can be included in this matrix. Table 2 shows an example of this type of matrix for a bioreactor and includes both the process parameters and the categorization from the risk analysis, as well as the impact the process parameters have, and the recommended studies to establish the control strategy.


Careful process characterization is important in the successful completion of pivotal manufacturing batches for Phase 3 trials to take place, and PPQ is an important step. The ICH guidelines serve as a high-level guideline,   however, there are many different approaches to the detailed planning of a PPQ master plan and scheduled GMP batches, and companies can build on their experience and be led by regulatory guidelines to enable the development of best practices.

Figure 1 gives a general timeline for the filing and approval of large molecules, condensing the points down into key outcomes and key deliverables. There are approaches to hasten development that may come with accelerated approvals and encouraging data from proof-of-concept clinical trials. Efforts to both understand the process, improve and validate methods, and beginning process characterization can be done in parallel in preparation for pivotal batches. A key consideration at this point is to avoid major changes if possible, and to begin stability studies as early as possible. Lessons learned from the process characterization, and the full understanding of the molecule and its critical attributes are the surest means of successful approval and lifecycle of the product. The earlier the process is confirmed, the more likely that changes will only be required if unexpected results and outcomes were to occur. Once all this is completed, primary stability drug substance and drug product batches can be made that will be used for the final process formulation and for fill–finish development. Other considerations to be made at this point will be the container, closure, and other aspects such as the requirement for a syringe.


Batches for stability testing will also be made, and at this point there are considerable time pressures, not least because a minimum of six months’ stability data is usually required for the biologics license application (BLA) filing. Regulatory bodies are looking to ensure the information on the process is all present and acceptable, so the data from the manufacturing batches, allied to the control strategy and the linkage of CQAs, will all be required for final approval. Ultimately, the solution requires a thorough understanding of the risks and what level of risk will be acceptable.

Conclusion
Process validation is a critical step of late-phase development in preparation for commercial launch. Navigating this step quickly and successfully requires integrated planning, timely execution and a deep knowledge of the process and molecule. A thorough understanding of both risk and commercial needs of the molecule also need to be taken into consideration during this pivotal time. Ultimately, it is important to maintain a line-of-sight from clinical development to commercial manufacturing with a view to patient impact. CP

References

1. https://www.fda.gov/downloads/drugs/guidances/ucm070336.pdf – Guidance for Industry – Process Validation: General Principles and Practices; FDA January 2011
2. https://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Quality/Q8_R1/Step4/Q8_R2_Guideline.pdf