A guide to EFPIA's Mock P.2 document

A favourable impact of implementing the new International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) quality principles and process analytical technology (PAT) to promote enhancements of manufacturing efficiency and quality is a strategic goal for both regulatory authorities and the industry.^1–3

Both industry and the public can benefit from ICH Q8 (pharmaceutical development), ICH Q9 (quality risk management [QRM]), and PAT principles and strategies as they foster quality by design (QbD) approaches during development of the process, and provide a platform for 'continuous improvement'.

Demonstrating adherence to modern science- and risk-based design, manufacturing, and modern control principles and strategies may allow authorities to adjust the level of regulatory oversight to the level of science-based' process understanding' and risk mitigation.

To support public discussion within the industry, and between industry and regulators leading towards this goal, European Federation of Pharmaceutical Industries Association (EFPIA) representatives have developed a discussion document (common technical document [CTD] section called 'Mock P.2'). The document discusses Examplain, a fictitious tablet product, which exemplifies the role of QRM and PAT to achieve QbD. This paper summarizes the main features of the Mock P.2 document.

Mock P.2 aims to:

• Promote discussion and learning between companies, and between EFPIA and regulatory authority reviewers and inspectors.

• Facilitate the scientific and regulatory dialogue between industry and regulatory authorities on the presentation of enhanced product and process understanding in regulatory dossiers.

• Exemplify application of QRM principles (ICH Q9) during the development process.

• Illustrating PAT-based manufacturing and process control strategies that enable 'continuous quality verification', and ultimately 'real-time release'.

• Demonstrating how application of QbD (ICH Q8) principles can be used to establish process understanding and the design space.

The design space is defined as the multidimensional combination and interaction of input variables (for example, material attributes) and process parameters that have been demonstrated to provide assurance of quality.¹ Working in the design space is not considered as a change. Movement out of the design space is considered to be a change and would normally initiate a regulatory post approval change process. Design space is proposed by the applicant and is subject to regulatory assessment and approval.

It is important to understand that the discussion document illustrates just one of many possible approaches to achieve the ultimate goal of QbD. Therefore, the document must not be taken as 'a standard recipe' for product development nor as a complete example for an actual submission document.

Products and manufacturing processes should be designed to manage variation. Formulation and process design in combination with risk management principles should be applied to ensure QbD is achieved during development and manufacture, yielding good understanding and control of variation. Manufacturing processes based on QbD are a means to mitigate the risk of variable product unit quality. Use of design space (ICH Q8 ), QRM (ICH Q9), and PAT principles, tools and practices, enables QbD to achieve the desired state.

The Mock P.2 submission

The Mock P.2 document gives examples of how quality can be built-in during formulation and process development. It includes the following concepts:

• The use of models and algorithms.

• The use of in-line and at-line tools.

• Using prediction models to establish the design space.

• Design space not requiring 'edge of failure' — a process parameter value that, if exceeded, means adverse effect on the process output or product quality.

• Linking control strategy to design space.

• The use of QRM principles.

In particular, it describes how enhancement of process understanding can be applied using prior knowledge and iterative application of QRM principles based on the target product profile during the design and development process.

The document is complemented by a 'mock' section P.3.3 excerpt outlining the process control strategy proposed for routine manufacture, which is based on the knowledge and understanding available and linked to a QRM review of all information available.

Figure 1 describes how formulation and process design are performed for the development of a 'simple' product. For simple formulations, it is typical that the manufacturing process and the formulation are developed first and then the process is optimized. Based on all the information available, and using QRM, it should be relatively straightforward to propose a design space — identifying what could be important to quality steps or parameters that require special attention or control. It should also be obvious what steps or parameters are critical to quality; that is, have an impact on quality of product supplied to the patient.

Figure 1 Iterative approach to establish design space and control strategy.

This analysis should also lead to a control strategy for application during initial routine manufacture. It is considered that this control strategy could change as more experience is gained during routine manufacture. Examplain, a fictitious tablet product, is a conventional, wet granulated tablet formulation of a relatively low dose, highly soluble, highly permeable (biopharmaceutical classification system class I) drug substance with low bulk density, which has some degradation potential. A wet granulation manufacturing process can be proposed based on a few experiments because of the low bulk density (Figures 2–3).

Figure 2 ´Examplain´ process scheme.

Monitoring the high-shear mixing process by power consumption or other techniques —such as acoustic spectroscopy —offers opportunities for better process understanding and advanced control of, for example, process end-points. Similarly, the fluidized bed-drying unit operation could be evaluated during development (Figure 4).

Figure 3 Granulation details.

Moisture content has been identified as a critical attribute of fluidized bed-drying — the critical manufacturing step.

Figure 4 Fluid-bed drying details.

The design space

Focusing on the evaluation of water content in tablets, the stepwise approach to establish a design space is described in detail in this section.

First, a QRM step was applied based on prior knowledge (Figure 5) to estimate what factors could be judged less important and what factors required further studies. From this evaluation factors to be included in the study were identified and introduced into a design of experiments (DOE) plan for formulation development including:

• drug substance particle size

• selection of other raw materials (lactose versus mannitol) and ranges, for example, magnesium stearate

• tablet hardness

• tablet dissolution (disintegration)

• evaluation of water content in tablets

• initial evaluation of content uniformity.

Figure 5 QRM approach.

An initial risk assessment identifying potential interactions between unit operations and key attributes is the starting point for the process development work. Risk assessment, control and communication should be frequently reviewed during the development work.

Figure 5 illustrates the result of single review, however, there should be multiple cycles to systematically reduce overall risk to the product, and build the overall control (risk mitigation) strategy.

All parameters relevant to wet granulation were identified from an Ishikawa diagram, and analysed using a detailed risk assessment failure mode and effects analysis (FMEA) approach to establish those process parameters associated with water content, the critical quality attribute.

Process understanding has been generated through prior knowledge, experimental development data (DOE studies at 1 kg and 25 kg scale) and use of multivariate experimental plans with in-line analytical measurements. Those interactions identified as critical to quality should be reflected in the control strategy versus interactions used as part of a monitoring strategy. Key granulation process variables are considered in DOE studies for Examplain (Table 1).

Table 1 Key process variables.

Within these studies, the relevance of mixing speed and water addition for disintegration properties of tablets was established.

The design space for water addition rate and water addition time is expressed as the acceptable area, with regard to preventing fines and degradation formation. An area of failure regarding disintegration is not found (Figure 6).

Figure 6 Design space for water addition rate and time.

As a consequence, the water addition rate and time can be adjusted within the design space to obtain desired granule properties with respect to flowability, compressibility, degradation and suitability as input to the next processing step: drying. In addition, multivariate models to predict disintegration could be applied, and the impact of scale changes can be understood.

Design spaces may be multidimensional and, therefore, represented in different, nonvisual ways. One example is using process trajectory charts as visual aids to illustrate the design space concept for fluid-bed drying. Process trajectories based on in-line measurement using near infrared (NIR) spectroscopy are established on a 1 kg scale and confirmed on a 25 kg scale (Figure 7).

Figure 7 Proposed design space for fluid bed drying as described by a trajectory of water content measured in-line using NIR spectroscopy as a function of time.

PAT-based manufacturing and control strategy

The granulation process is controlled by measuring the power consumption during granulation. The amount of water per mass of granulate, and the addition feed rate, are fixed based on process understanding from DOE. The power consumption peak is the starting point for 6 min time interval to finish wet massing. This accommodates variable raw material properties (for example, water content, particle size distributions). The wet granulation design space is part of the control strategy to provide additional assurance of satisfactory granulation.

PAT as an enabler for QbD can be described by a set of tools (multivariate tools for design, data acquisition and analysis, process analysers, process control tools, and continuous improvement and knowledge management tools) and principles (risk-based and integrated approach, real-time release). These principles and tools are applied to the Examplain mock example, and are illustrated in Figure 8. The control strategy contains some elements of redundancies that may be reviewed and adjusted when more manufacturing experience is available.

Figure 8 Summary of Examplain manufacturing and control concept.

Conclusion

The Examplain Mock P.2 document provides an example to build formulation and process understanding in a submission. It gives an example of how ICH Q8, Q9 and PAT can be used to enable QbD leading to significant regulatory flexibility, including real-time release and postapproval continuous improvement.

As a means to promote and structure discussions on how these concepts can be implemented into a market application submission, EFPIA's Mock P.2 document illustrates how:

• Design space can be established based on QbD, process understanding and risk management.

• Real-time release approaches can be justified based on process control and monitoring schemes, which are designed by applying process understanding and consequent iterative application of QRM tools.

• Conventional process validation approaches can be replaced by continuous process verification.

• Postapproval changes to site and scale can be anticipated in the established design space by incorporating factors of equipment, scale and site and linkage to the routine control strategy.

The case study of Examplain tablets facilitates detailed discussion of implementation concepts regarding:

• Continuous improvement within the design space.

• Real-time release of batches that is based on in-process information of critical material attributes, and predictive modelling.

• Continuous quality verification on a batch-by-batch basis to supersede conventional approaches to process validation. Specifically, validating a successful testing of predefined numbers of batches is redundant.

• Design space that encompasses aspects of scale, environmental aspects of site, packaging, as well as final product stability. This is based on evidence that process performance is unrelated to stability behaviour of the modulation.

Chris Potter and Staffan Folestad both at AstraZeneca

Rafael Beerbohm at Boehringer Ingelheim

Gordon Muirhead at GlaxoSmithKline

Stephan Roenninger at F. Hoffmann La Roche

Alastair Coupe and Alistair Swanson both at Pfizer

Fritz Erni at Novartis

Gerd Fischer at Sanofi-Aventis

References

1. ICH Q8 ''Pharmaceutical Development'', Step 4 Document, November 2005.

2. ICH 09 "Quality Risk Management", Step 4 Document, November 2005.

3. FDA Guidance for Industry "PAT — A Framework for Innovative Pharmaceutical Development, Manufacturing and Quality Assurance", September 2004.