Quality by design (QbD) is a systematic and proactive approach to integrate quality into the biopharmaceutical manufacturing process from its inception, aiming to reduce product variability and defects, enhance development and manufacturing efficiencies, and improve overall product life cycle management. By establishing a well-defined design space and employing robust process controls, QbD can help minimize variability in production and enhance process efficiency. This methodology not only facilitates a deeper understanding of product and process interdependencies but also allows for scalability and flexibility within the manufacturing process. Regulatory compliance is aligned with FDA and ICH guidelines under QbD, which emphasizes a science- and risk-based approach in addition to detailed documentation and proactive quality measures, thus potentially expediting regulatory reviews and approvals such as IND and BLA filings. Moreover, the adoption of important tools such as process analytical technology (PAT), design of experiments (DOE), and continuous process verification underpins ongoing improvements and sustains the adaptability of manufacturing operations. Conversely, reliance solely on end-product testing poses risks including increased variability, regulatory scrutiny, and escalated costs. QbD’s comprehensive approach and clear objectives not only meet but can anticipate regulatory requirements, establishing a new standard in biopharmaceutical manufacturing that prioritizes quality at every step.
Emphasizing Intentional Quality in Biomanufacturing Biopharmaceutical manufacturing processes are inherently complex, and even minor adjustments in process parameters can significantly alter the properties and quality of the resulting drug substances and drug products. Traditionally, due to limited mechanistic understanding of cell culture and fermentation, processes were rigid to maintain consistency, and quality was only verified after product manufacturing. This conventional approach lacks the flexibility to adapt to changes in raw material sourcing or advancements in process technology, involves extensive hold times for products during release testing, and incurs high costs when batches fail to meet specifications.1
A more effective strategy is to develop comprehensive product and process knowledge early on, enabling the integration of quality into the design of biopharmaceutical products and manufacturing processes from the outset. Regulatory agencies have acknowledged the shortcomings of end-point testing and advocate for designing quality into products early on to better meet patient needs while ensuring manufacturability at the required safety and efficacy standards. Incorporating quality at each step of the process also enhances overall performance and positively influences subsequent steps.2
Simultaneously, a deeper understanding of manufacturing processes facilitates the development of robust yet adaptable strategies capable of managing common variabilities within the bioprocessing environment, significantly reducing the time to product release, minimizing the risks of batch failures and the need for extensive post-production processing.3
The Quality-by-Design Concept
This proactive approach of designing quality into pharmaceutical products and processes from the first stages of development is known as Quality by design (QbD). Its foundation lies in the Pharmaceutical Quality for the 21st Century initiative launched in 2002 by the US Food and Drug Administration (FDA) to explore approaches to reducing risk and enhancing quality across the drug development spectrum.3
The principles of QbD are detailed in various guidelines of the International Council for Harmonization (ICH) Technical Requirements for Registration of Pharmaceuticals for Human Use, including Q8,4 Q9,5 Q10,6 and Q11,7 which focus on pharmaceutical development, quality risk management, quality systems, and development and manufacture of drug substances, respectively. These guidelines also include ICH E68 and ICH E8,9 which describe good clinical practice (GCP) and general considerations for clinical trials, respectively, reflecting a comprehensive framework for ensuring consistent production of high-quality products tailored to meet patient needs.10,11
The QbD approach begins with the engineering of the biologic drug substance to possess specific properties that assure safety, efficacy, and stability, facilitating both storage and handling while optimizing manufacturability and delivery in alignment with patient population needs and desired clinical outcomes and product performance.10,12 Early development stages include rigorous risk assessments to identify and mitigate potential variabilities that could impact product quality. This is followed by a detailed exploration of process parameters to understand their effects on critical quality attributes, enabling the development of a process that maximizes yield and purity while ensuring safety.
To mitigate identified risks, process controls are implemented wherever feasible, complemented by continuous monitoring to enhance process capability and inform future product development strategies.1,2,10–12 As the product and process mature toward commercialization, this ongoing accumulation of knowledge supports iterative improvements and robustness of the manufacturing process.
Noteworthy milestones in the application of QbD include the approval of Perjeta (pertuzumab) in Europe in March 2013 for HER2+ breast cancer — marking the first European QbD approval for a monoclonal antibody.12 This was closely followed by the U.S. approval of Gazyva (obinutuzumab) for follicular lymphoma in November 2013, representing the first complete QbD submission in the United States.13 These approvals underscore the global recognition and implementation of QbD principles in pharmaceutical manufacturing.
Process and Product Understanding
The primary goal of embedding quality into biomanufacturing is to fulfill patient needs by delivering safe, effective, compliant, and cost-efficient medicines. Achieving these objectives hinges on a profound understanding of patient requirements and product and process knowledge. Initially, this information helps establish the quality target product profile (QTPP), which then guides the identification of critical quality attributes (CQAs), enabling the design of a manufacturing process that consistently delivers a drug product with the desired product quality.
For monoclonal antibodies (mAbs), the QbD journey begins with the design and engineering of the biologic candidate to align with the QTPP, targeting desired outcomes in efficacy, safety, manufacturability, stability, and overall quality.10 This phase includes various activities, such as sequence design, cell line engineering, isotype selection, posttranslational modifications, and humanization, among others.
Gathering product knowledge involves leveraging diverse data sources, including structural and physicochemical assessments, understanding the therapeutic mechanism of action, and analyzing the pharmacokinetics and pharmacodynamics of the drug product.10,14 This also encompasses insights from clinical and nonclinical studies, experiences with similar molecules, and other pertinent publicly available information. For mAbs, CQAs typically include attributes related to structural integrity, such as correct folding, disulfide bond formation, posttranslational modifications such as glycosylation patterns, and aggregation propensity.14
Simultaneously, process knowledge is acquired by meticulously exploring how various process parameters affect the quality attributes. Parameters that significantly influence CQAs are recognized as critical process parameters (CPPs). Achieving a complete understanding of the processes, especially in cell culture and fermentation, is challenging owing to the complexity and interdependence of the various process parameters and the substantial impact even minor changes can have on product quality. Therefore, it is crucial to map out the interactions between these parameters to understand how variations can affect product quality.14 This comprehensive mapping also involves identifying potential variability sources and assessing the risks they pose to the CQAs.
Establishing the Design Space Using Design-of-Experiment Techniques
In biomanufacturing, potential variabilities can arise from numerous sources, such as raw materials and consumables, operational inputs (e.g., measurements, methods, personnel, equipment), and environmental conditions. Identifying these sources and assessing their potential impact on CQAs is the first step in controlling variability. Techniques like failure mode and effects analysis (FMEA) are commonly employed to evaluate the likelihood of an occurrence and the risk associated with each source of variability.14
Following risk assessment, design of experiments (DoE) is implemented to investigate the impacts of these high-risk variables. This approach allows for the simultaneous exploration of multiple process parameters, enhancing the efficiency of the study by minimizing the time and costs associated with traditional one-factor-at-a-time (OFAT) methods.15 DoE provides valuable insights into the interrelationships between process parameters and helps identify the CPPs that need monitoring and control to maintain consistent product quality.
The data obtained from DoE studies are analyzed using multivariate techniques to define a design space — a range of conditions that specifies the limits for CPPs ensuring optimal product quality. This design space is applied across all stages of manufacturing, including cell culture, harvest, chromatography, ultrafiltration/diafiltration, viral filtration, formulation, and final fill. It also extends to cell clone and media selection, as well as evaluating the impacts of process scale-up.9 Given that these studies typically utilize scale-down models, verifying that these models accurately represent large-scale conditions is crucial.15 The design space is not static but evolves based on ongoing data collection and analysis throughout the development and commercial manufacturing stages, allowing for continual refinement and optimization of the process.
Process Monitoring and Control
Once the design space for a biomanufacturing process is established, the focus shifts to implementing measures to ensure consistent operation within this defined space. This involves setting precise specifications and testing requirements for raw materials, as well as establishing conditions for storage and shipping to safeguard the integrity of materials before they enter the process.14
Within the process itself, various methods are employed to monitor and control operations.14,15 The use of process analytical technologies (PATs), ideally inline tools, offer real-time data on CQAs, enabling continuous monitoring. Automation solutions equipped with feedback loops play a crucial role in maintaining process control by dynamically responding to any variations in CQA values, ensuring sustained product quality. Additionally, alarms are set to alert operators when CQA values approach the boundaries of the design space, signaling the need for potential intervention.
Statistical process control (SPC) techniques, which utilize statistical methods to manage and control processes, are integral to this monitoring framework. Predictive algorithms further enhance process control by allowing for adjustments in downstream operations based on upstream data, ensuring a proactive approach to quality management.
These in-process controls are complemented by end-product release testing, which verifies the quality against established specifications. The control strategy, much like the design space, is dynamic and continually refined in response to new data and insights gained as the process progresses from development to commercialization.15
The complexity of biologics and cell culture/fermentation processes makes controlling biopharmaceutical manufacturing particularly challenging compared with small molecule drug production.15 Biologic entities, such as recombinant proteins and antibodies, are inherently heterogeneous, requiring a suite of analytical techniques for comprehensive characterization. Their complex structures must be preserved during storage and shipping, often necessitating cold-chain logistics and meticulous monitoring using sophisticated bioassays. Moreover, given that nearly all biologic drug products are delivered parenterally, stringent microbial and viral control measures are essential to ensure sterility and safety.
Ongoing Improvement
The principle of QbD emphasizes not just the initial establishment of a robust biomanufacturing process with a well-defined design space and consistent performance but also the imperative of continuous improvement.14,15 This ongoing enhancement process is fueled by the continuous generation of new data throughout the development phases and following the commercialization of a biologic drug. It also relies heavily on collaborative efforts among various stakeholders involved in bioprocessing, including academic researchers, technology suppliers, contract manufacturers, and other biopharmaceutical entities.
Continual refinement of the original process design space is informed by a variety of data sources. This includes results from clinical trials, stability studies, changes in the process, assessments of comparability, and the integration of new analytical technologies.16 These insights often prompt the need to adjust the design space for specific drug substances or products to accommodate new findings or improvements in platform processes. Another critical aspect of ongoing improvement is the continuous process verification performed during commercial production. This verification ensures that the control strategy remains effective and adapts to any new challenges or information, thus maintaining the integrity and quality of the product throughout its life cycle.
Data Management and Processing Challenges
Implementing QbD in biomanufacturing is fundamentally a data-driven strategy that necessitates comprehensive access to both the tools and expertise required for managing and processing substantial data volumes. It is crucial that the software supporting QbD activities not only be user-friendly for operators but also capable of delivering pertinent analysis results swiftly and in formats that are easy to interpret.1
SPC and statistical quality control (SQC) systems help understand the variability in biopharmaceutical manufacturing processes and if presented appropriately can enable operators to clearly visualize when processes are deviating from desired performance. Similarly, multivariate data analysis software solutions must present CPP and CQA correlation data in an easy-to-read format that clearly delineates the CPPs with an impact on CQAs.
Equally important are modeling solutions, such as multivariate stochastic process control tool sets, which can predict potential future deviations based on current trends. These predictive models are pivotal for preemptive process adjustments, ensuring consistent product quality and reducing the likelihood of costly disruptions.
Several Keys to Success
Successfully implementing a QbD approach in biologic drug production is a complex undertaking that requires substantial investments in terms of financial resources, human capital, and time.17 The benefits of QbD are manifold, but realizing these benefits hinges on several critical prerequisites:
Leadership commitment: The foundation for successful QbD implementation is a strong commitment from senior management. Active leadership involvement is crucial for fostering a quality culture that fully embraces the principles of QbD.
Access to tools and integration: Effective implementation also depends on access to the right tools, technologies, and processes. These components must not only be advanced but also seamlessly integrated with each other and into daily operational workflows to facilitate smooth functioning.
Comprehensive preparation: This includes thorough research on QbD implementation strategies, extensive training for employees, and the meticulous development of detailed plans and strategies to guide the implementation process.
Central to all these elements is a robust quality management system (QMS) tailored for biopharmaceutical operations. A well-designed QMS supports the integration of quality into biomanufacturing processes from the start.2 It enables efficient electronic document management, ensures comprehensive employee training, automates workflows, and provides robust mechanisms for problem resolution and corrective actions tracking. Importantly, the QMS must be implemented across the entire organization, ensuring uniformity in practice regardless of location or role. This uniformity facilitates effective data sharing across different products and processes, enhances decision-making, fosters the development of more robust processes, and supports ongoing improvement initiatives.
Many Benefits of Designing Quality into Biomanufacturing
Integrating quality into the development of biopharmaceutical processes and products from the outset offers substantial benefits beyond the fundamental assurance of quality, safety, and efficacy.2,11 A deep understanding of product and process combined with a well-defined design space results in more efficient processes that can be effectively scaled up, encountering fewer production issues and reducing the need for reprocessing steps. Overall, this approach leads to shorter development timelines and lower costs.
The regulatory advantages of adopting a QbD approach are also significant.14 Manufacturers can expedite regulatory reviews for Investigational New Drug (IND) and Biologics License Application (BLA) submissions by demonstrating thorough understanding of the product and its manufacturing process and proactively identifying and mitigating potential risks, thus instilling greater confidence in regulatory agencies regarding the robustness of manufacturing methods. Moreover, postapproval change management becomes streamlined under a QbD approach, where the associated documentation provides regulatory agencies with a detailed insight into the development and manufacturing journey, fostering trust and collaborative interactions with regulatory agencies. Processes that are continuously monitored and controlled using PAT for real-time data collection and verification are well-positioned to transition to real-time release testing (RTRT), offering further efficiencies and regulatory flexibility.
Conversely, relying solely on end-product testing as opposed to embedding quality early in the biomanufacturing process carries several risks.11 While initially appearing to lower financial exposure for early-stage projects, this approach can lead to inflexible processes that produce inconsistent results. Minor variations in raw materials, equipment status, or process conditions can significantly impact the quality of the final product, complicating clinical trials, scaling efforts, and ultimately leading to project delays, increased regulatory scrutiny, and higher costs.
References
1. “Quality by Design in Biomanufacturing: Implementing real-time data consolidation, visualization and modeling for improved quality and lower compliance costs.” Emerson. Feb. 2020.
2. Love, Ave. “5 Key Elements of Quality by Design in Pharma.” MasterControl GxP Lifeline. 16 Aug. 2022.
3. Pharmaceutical Current Good Manufacturing Practices for the 21st Century: A Risk-Based Approach; Establishment of a Public Docket. U.S Food and Drug Administration. 27 Feb. 2003.
4. ICH Q8 (R2), Pharmaceutical development. International Council for Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use. Aug. 2009.
5. ICH Q9 (R1), Quality Risk Management. International Council for Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use. Jul. 2023.
6. ICH Q10, Pharmaceutical Quality System. International Council for Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use. Jun. 2008.
7. ICH Q11, Development and manufacture of drug substances (chemical entities and biotechnological/biological entities). International Council for Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use. Nov. 2012.
8. ICH E6 (R3), Guideline on Good Clinical Practice (GCP). International Council for Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use. May 2023.
9. ICH E8 (R1), Guidelines on General Considerations for Clinical Studies. International Council for Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use. Oct. 2021.
10. Arora, Taruna et al. “Quality by design for biotechnology products—part 1.” BioPharm International. 29 Dec. 2013.
11. De La Madrid, Patricia. “Quality By Design in Biologic Drug Development.” 5Biologics Blog. 13 Jun. 2022.
12. Luciani, Francesca et al. “Implementing quality by design for biotech products: Are regulators on track?” MAbs. 7: 451–455 (2015).
13. Zettl, F. “Development and Implementation of a Design Space for a New Therapeutic Antibody: A Case Study.” IBC Biopharmaceutical Development and Production Conference. San Diego, CA, 24–27 Mar. 2014.
14. Kalghatgi, Sameer. “Quality by Design (QbD) in Biomanufacturing: A Comprehensive Guide.” LinkedIn. 27 Oct. 2023.
15. Anliker, Sally et al. “Quality by Design for Biotechnology Products—Part 2.” BioPharm International. 22: 40-45 (2009).
16. Anliker, Sally et al. “Quality by Design for Biotechnology Products—Part 3.” BioPharm International. 23: 40-45 (2010).
17. Thacker, Molly. “3 key ingredients for winning at Quality by Design in pharma.” Montirum Blog. Aug. 2022.