Quality by Design in Biopharmaceutical Development

Quality by Design (QbD) is a systematic, risk-based, and science-driven approach to pharmaceutical development and manufacturing. It emphasizes a shift from empirical product testing to knowledge-based design and control, ensuring that quality is built into the product from the outset, not merely tested at the end.

1. Introduction to Quality by Design (QbD)

1.1 Origins of QbD

The foundational concept of Quality by Design was introduced by Dr. Joseph Moses Juran (1904–2008), a Romanian-born Jewish-American engineer and management consultant, widely regarded as one of the fathers of modern quality management. Juran was the first to formalize the idea that “quality should be designed into a product, not inspected in.” He proposed that every process should begin with a clear understanding of customer needs, and that variation and inefficiency should be minimized through intelligent design rather than correction.

Though Juran’s ideas were initially developed in the context of post-war industrial engineering and management, they were later adopted into pharmaceutical regulation through the International Council for Harmonisation (ICH). His legacy underpins much of the QbD philosophy enshrined in ICH Q8–Q12, as well as broader FDA and EMA lifecycle management strategies.

Without a standard, there is no logical basis for making a decision or taking action.Joseph M. Juran

1.2 Overview of the QbD Process

QbD integrates product development, process understanding, risk control, and regulatory lifecycle strategy into a single coherent framework:

Modality-specific considerations

While the QbD framework is applicable across all product classes, its implementation must account for the inherent differences in molecular complexity, manufacturing platforms, and clinical use cases:

In this guide, each QbD step is illustrated with examples from all three modalities to provide broad applicability while maintaining scientific precision.

2. Global Regulatory Framework for QbD

Quality by Design is globally supported by a well-established regulatory framework that combines harmonized ICH guidelines, binding national laws, and interpretive guidance from regional health authorities. These form the scientific and legal basis for implementing QbD across the product lifecycle.

2.1 Core International Guidelines (ICH)

2.2 United States (FDA)

2.3 European Union (EMA)

2.4 Japan (PMDA / MHLW)

2.5 China (NMPA / CDE)

2.6 Canada (Health Canada)

2.7 India (CDSCO)

2.8 Industry Best Practices (PDA & USP)

These regulatory and technical sources collectively empower pharmaceutical developers to design robust, science-driven, and risk-managed processes that not only meet but anticipate regulatory expectations across markets.

3. QbD Principles Across Modalities

The implementation of QbD varies across product modalities due to differences in manufacturing complexity, biological variability, and the maturity of platform knowledge. However, the underlying principles, product and process understanding, risk management, and control,remain constant.

Small Molecules

Development of small molecules follows deterministic chemical synthesis pathways. This allows for extensive process control and predictability.

Biologics and Oligonucleotides

Biological products exhibit inherent variability due to the use of living systems. QbD in biologics focuses on understanding and controlling sources of biological and process variability.

Cell and Gene Therapies (CGTs)

CGT manufacturing is less defined, typically patient-specific, and presents unique challenges due to variability, low yields, and limited in-process testing.

QbD in CGTs emphasizes a comprehensive understanding of Critical Material Attributes (CMAs), flexible control strategies, and platform analytics for rapid process iteration.

4. Target Product Profile (TPP) and QTPP

A robust Quality by Design framework begins with the Target Product Profile (TPP) and Quality Target Product Profile (QTPP). These foundational tools serve as strategic, preclinical-to-commercial planning instruments that define the desired clinical outcomes and translate them into quality expectations for product and process development.

The TPP aligns product development with therapeutic intent, regulatory strategy, and patient needs. The QTPP, derived from the TPP, delineates the quality attributes essential to safety, efficacy, and performance of the final product. Together, these profiles shape CQA selection, analytical development, specification setting, and control strategy design.

4.1 Example A, Allogeneic Cell Therapy (Renal Indication)

4.2 Example B, Biologic (Monoclonal Antibody, Oncology Indication)

4.3 Example C, Small Molecule (Oral Antiviral, Hepatitis C)

4.4 Strategic Role of TPP and QTPP in QbD

The TPP defines the product’s clinical and patient-focused objectives, guiding CMC decisions from the outset. From the TPP, the QTPP is constructed by identifying attributes relevant to safety, efficacy, and stability; translating these into measurable quality targets; and aligning them with regulatory and pharmacopoeial requirements.

The QTPP is then used to:

• Guide the selection of Critical Quality Attributes (CQAs).
• Define Analytical Target Profiles (ATPs) and method requirements.
• Justify specification limits during regulatory submission.
• Drive process understanding and design space decisions.

5. Critical Quality Attributes (CQAs)

Critical Quality Attributes are defined by ICH Q8(R2) 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.ICH Q8(R2)

CQAs directly link the Quality Target Product Profile (QTPP) to the product’s design and control strategy. They are identified based on their impact on safety, efficacy, stability, and performance and are tightly linked to regulatory specifications and clinical outcomes.

5.1 Principles of CQA Identification

CQA identification is a risk-based and science-driven process. It involves evaluating potential attributes from the QTPP, process knowledge, and prior data; applying risk assessment tools (e.g., Failure Mode and Effects Analysis [FMEA], Parameter–Attribute Matrix [PAM]); and determining which attributes are critical, non-critical, or key. Key criteria for assigning CQA status:

1. Clinical Relevance: Does variation in the attribute impact safety or efficacy?
2. Process Sensitivity: Is the attribute sensitive to changes in process parameters?
3. Detectability and Control: Can it be measured and controlled with appropriate precision?

ICH Q8, Q11, and Q14 emphasize that this process is iterative, evolving through development as product and process understanding increases.

5.2 Modality-Based Examples of CQAs

5.3 Lifecycle Alignment of CQAs

CQAs evolve in importance and definition through the development lifecycle:

5.4 Modality Spotlight: Attribute-Specific Considerations

Small Molecules.

• Dissolution is often a surrogate for bioavailability and thus a critical performance CQA.
• Polymorphism must be tightly controlled due to its effect on solubility and stability.
• Residual solvents are managed via ICH Q3C thresholds, with limits defined by Class 1–3 toxicity.

Biologics / Oligonucleotides.

• Glycosylation patterns affect efficacy (e.g., ADCC) and immunogenicity.
• Aggregate levels are monitored due to links with immunogenicity and reduced activity.
• Truncated sequences or backbone modifications in oligos must be quantified to ensure safety.

Cell and Gene Therapies.

• Viability is essential for efficacy and is often a specification CQA in cell therapies.
• Identity via flow cytometry ensures the presence of therapeutic cell types (e.g., CD19+ CAR-T).
• Transduction efficiency and empty/full capsid ratio affect gene therapy potency and toxicity.
• Sterility and endotoxin control is mission-critical due to the route of administration (often IV).

5.5 Integration with Control Strategy

Once identified, CQAs must be mapped to appropriate analytical methods (per ICH Q2/Q14); specifications aligned with regulatory guidance; control strategies that maintain each CQA within its proven acceptable range (PAR); and real-time release or trending programs in the commercial lifecycle. For high-risk CQAs, in-process controls (IPCs) and online PAT tools are often justified to ensure real-time assurance of quality.

6. Risk Assessment and Management

A foundational principle of QbD is the proactive identification, assessment, control, and review of risk to ensure product quality. This risk-based approach is enshrined in ICH Q9 and is reinforced throughout ICH Q8–Q12 as essential for robust development, lifecycle management, and regulatory decision-making. Risk assessment is not a one-time exercise, but rather a dynamic, iterative process that evolves as product and process knowledge matures.

6.1 Objectives of Risk Assessment in QbD

• Identify critical factors affecting product quality (e.g., CPPs, CMAs, CQAs).
• Prioritize development activities (e.g., DoE focus, PAT deployment).
• Design control strategies proportional to the level of risk.
• Enable regulatory flexibility via science- and evidence-based justification (e.g., design space, PACMPs).

Effective risk assessment ensures that resources are allocated efficiently and that quality risks are mitigated before they manifest as deviations or failures.

6.2 Key Tools for Risk Assessment

These tools can be used individually or in combination, depending on the development phase, data availability, and risk profile.

6.3 Risk Assessment Process Flow (Aligned with ICH Q9)

This framework supports continuous improvement and aligns with ICH Q10 (Pharmaceutical Quality System) and ICH Q12 (Lifecycle Management).

6.4 Modality-Specific Risk Considerations

For complex modalities, risk assessments must also encompass donor-to-donor variability (CGT); vector manufacturing inconsistencies (AAV/LV); and assay variability in cell-based or bioassays (Biologics/CGT).

6.5 Risk-Based Approach to Control Strategy Design

Effective risk assessment enables a tiered control strategy:

This stratified approach is consistent with PDA TR 60, TR 81, and the ICH Q12 Enhanced Approach, where higher understanding justifies more flexible regulatory commitments.

7. Design of Experiments (DoE) and Control Strategy

Design of Experiments is a structured, statistical approach for investigating and understanding the relationship between input factors (e.g., process parameters, formulation variables) and output responses (e.g., CQAs, yield, titer). DoE is a core enabler of QbD, facilitating identification of Critical Process Parameters (CPPs), Proven Acceptable Ranges (PARs), and, ultimately, the Design Space. ICH Q8(R2) emphasizes the use of DoE as part of an “enhanced development approach,” distinguishing it from empirical or trial-and-error methods.

7.1 Objectives of DoE in QbD

• Identify CPPs and CMAs that impact product quality.
• Optimize multivariate relationships between inputs and CQAs.
• Quantify interaction effects and define Design Space.
• Support control strategy development with robust process knowledge.

DoE contributes directly to the scientific justification of specifications, in-process controls, and real-time release testing (RTRT), enabling flexibility in regulatory submissions under ICH Q8–Q12.

7.2 Common Types of DoE and Their Applications

7.3 Integration into Control Strategy

• Raw Material Controls (CMAs): Define source, acceptance criteria, and testing (e.g., FBS, vector plasmids, solvents).
• Process Controls (CPPs): In-process parameter limits (e.g., bioreactor pH, mixing time, shear rate).
• IPC/Release Specifications (CQAs): Acceptance criteria for release testing (e.g., purity, potency, AUC).
• Analytical Control Strategy (ATP–CMP–CMA): Based on method robustness and lifecycle (see ICH Q14).
• Real-Time Release Testing (RTRT): Enabled through PAT and validated predictive models.
• Proven Acceptable Ranges (PARs) and NORs: Defined ranges for operating within or outside the Design Space.
• Established Conditions (ECs): Legally binding control elements documented in regulatory submissions (per ICH Q12).

7.4 Modality-Specific Examples

DoE allows identification of interactions that would otherwise be missed in One-Factor-at-a-Time (OFAT) approaches, enabling more efficient, higher-confidence process development.

7.5 DoE and Design Space

A key regulatory benefit of DoE is that it enables registration of Design Space, which allows changes within its bounds without requiring regulatory resubmission (as per ICH Q8):

Movement within the Design Space is not considered a change and does not require regulatory approval.ICH Q8(R2)

DoE defines:

• Design Space (Multivariate): Justifies flexibility in operation and regulatory commitment.
• Normal Operating Ranges (NOR): Practical ranges for routine manufacturing within the Design Space.
• Proven Acceptable Ranges (PARs): Experimentally validated ranges for CPPs outside Design Space (optional).
• Control Space: Subset of the Design Space where operations are locked for regulatory or technical reasons.

7.6 DoE and Analytical Development

With the implementation of ICH Q14, DoE also plays a pivotal role in defining the Analytical Method Operational Design Region (MODR); identifying Critical Method Parameters (CMPs); and supporting lifecycle robustness and change control for analytical procedures.

8. Process Analytical Technology (PAT)

Process Analytical Technology is a cornerstone of the QbD paradigm, enabling real-time or near-real-time measurement and control of critical process parameters (CPPs) and critical quality attributes (CQAs). Unlike conventional quality testing performed post-batch, PAT ensures that quality is designed into the process and continuously monitored throughout manufacturing.

PAT is a system for designing, analyzing, and controlling manufacturing through timely measurements… to ensure final product quality.FDA

The goal is to reduce variability, enable process understanding, and support Real-Time Release Testing (RTRT) where feasible.

8.1 Regulatory Basis and Expectations

PAT implementations should be supported by risk assessment, method validation or model verification, and robust data infrastructure per GMP data integrity standards.

8.2 PAT Technologies by Modality

These tools allow for inline (direct), online (continuous sampling), or at-line (near process) analysis, depending on technical feasibility and regulatory acceptance.

8.3 Examples of PAT Implementation

8.4 Benefits and Challenges of PAT

To overcome these, industry best practices include early inclusion of PAT in QTPP and control strategy planning; use of modular validation plans aligned with risk-based ICH Q9 principles; and model lifecycle controls per ICH Q14, including method robustness and change management.

8.5 PAT in the Control Strategy Framework

PAT supports a hierarchical control strategy that can reduce or eliminate traditional end-product testing when predictive models are validated:

8.6 Digitalization and Future Outlook

Modern PAT systems are increasingly integrated with manufacturing execution systems (MES) and AI/ML-based analytics to support Multivariate Process Control (MVPC); Digital Twins for Bioprocessing; Predictive Maintenance and Automated Alarms; and Regulatory Submissions via Real-Time Data Repositories.

Regulators such as the FDA Emerging Technology Team (ETT) and EMA Innovation Task Force actively encourage dialog on PAT deployment in novel platforms, especially for continuous manufacturing and advanced therapies.

9. Design Space and Lifecycle Management

Design Space is a central concept in QbD, describing a scientifically justified, multidimensional range of input variables and process parameters that consistently result in a product meeting predefined quality criteria. It is a key enabler of regulatory flexibility, lifecycle efficiency, and robust control strategy implementation.

Working within the design space is not considered a change.ICH Q8(R2), Section 4.3

In a well-structured QbD system, the design space evolves from accumulated process knowledge, and supports predictable performance, efficient change management, and regulatory transparency across the product lifecycle.

9.1 QbD Spatial Framework

The QbD approach defines several interrelated conceptual spaces:

Understanding how these spaces interrelate supports both process performance and regulatory lifecycle management.

9.2 Establishing the Design Space

Design space is built progressively through:

1. Prior Knowledge: Platform experience, published data, similar molecules.
2. Risk Assessment: FMEA, PAM, Ishikawa, MVDA.
3. DoE Studies: Multivariate analysis of CPP–CQA relationships.
4. Scale-down Models: Lab and pilot scale data to define performance ranges.
5. Process Modeling: Mechanistic, statistical, or hybrid digital twins.

The design space must be experimentally justified and statistically robust. Its definition may include combinations of input material characteristics (e.g., particle size, media osmolality); process parameters (e.g., mixing speed, temperature, pH); and equipment attributes (e.g., impeller type, filter pore size).

9.3 Modality-Specific Examples

9.4 Lifecycle and Change Management: ICH Q12 Framework

ICH Q12 establishes tools to manage post-approval changes without resubmission for minor variations:

This framework promotes global regulatory alignment, reduces delays, and supports continuous improvement.

9.5 Control Strategy Alignment

The design space is directly linked to the control strategy, which defines how process variability is managed to maintain CQAs:

Movement within the design space is permissible without regulatory reporting, enabling operational agility and supply chain resilience.

9.6 Continuous Process Verification (CPV) and Design Space Reassessment

Design space does not end at marketing approval. It must be supported by ongoing verification: Annual Product Review (APR/PQR) trending; Statistical Process Control (SPC) of CPPs and CQAs; deviation and excursion analysis; and model maintenance and revalidation (for PAT/RTRT). Reassessment may be needed after facility transfer, scale changes, raw material source changes, or regulatory feedback or product complaints.

10. Case Studies in QbD Implementation

Real-world application of QbD demonstrates how a systematic, data-driven approach improves product robustness, regulatory acceptance, and manufacturing reliability. The following hypothetical case studies illustrate how QbD tools, such as Design of Experiments (DoE), multivariate data analysis (MVDA), Process Analytical Technology (PAT), and risk assessment,are applied across modalities to solve development and manufacturing challenges.

10.1 Small Molecule API, Polymorphic Control

Background. A parenteral cyclodextrin therapeutic was designed to sequester a toxic metabolic byproduct via inclusion complexation. In early batches, variability in side-chain substitution led to inconsistent binding affinity and solubility.

QTPP element. Predictable complexation efficiency; clear, soluble injectable; safe for bolus or central line use.

CQA identified. Degree and uniformity of substitution (DS), solubility, complexation performance.

QbD approach.

• FMEA flagged reagent ratios and pH during derivatization as high-risk.
• DoE optimized molar excess, temperature, and reaction time.
• NMR and HPLC used to quantify DS consistency.

Outcome.

• Control space established for side-chain modification reaction.
• In-batch variability reduced by >70%; complexation performance stabilized across lots.

>70%

reduction in in-batch variability

DS

registered as an Established Condition

10.2 Biologic (Monoclonal Antibody), Aggregate Reduction

Background. A monoclonal antibody exhibited high aggregate levels (>4%) during purification, compromising product stability and shelf life.

QTPP element. High stability over 24 months; minimal aggregation for safety and efficacy.

CQA identified. Aggregates, charge variants, bioactivity.

QbD approach.

• MVDA used to correlate process history with aggregation trends.
• DoE applied to optimize elution pH and buffer molarity in protein A and CEX chromatography.

Outcome.

• Robust Design Space defined.
• Aggregation reduced by 40% across 3 batches.
• Control space defined for elution conditions.
• Enabled shelf-life extension to 30 months.

>4%

initial aggregate level

40%

aggregation reduction across 3 batches

30 mo

extended shelf life

10.3 Gene Therapy (AAV Vector), Titer and Empty/Full Ratio Control

Background. AAV-based gene therapy for retinal disorder faced batch failures due to inconsistent vector genome titer and high empty/full capsid ratios.

QTPP element. Reliable delivery of transgene; consistent genome copy per dose.

CQA identified. Genome titer, empty/full ratio, capsid integrity.

QbD approach.

• Ishikawa diagram used to map sources of titer variability.
• PAT (qPCR and UV at-line) integrated into UF/DF and anion exchange steps.
• DoE on salt gradient profiles.

Outcome.

• Lot rejection rate reduced from 30% to <5%.
• Real-time titer trending improved process control.
• ECs defined for purification parameters.

30%

prior lot rejection rate

<5%

lot rejection rate after QbD

10.4 Cell Therapy (Allogeneic Renal Repair), Viability and Dosing Uniformity

Background. An allogeneic cell therapy targeting CKD showed inconsistent post-thaw viability and dose delivery, impacting efficacy.

QTPP element. Viable, potent dose with single-administration intent.

CQA identified. Post-thaw viability, cell count per dose, residual DMSO.

QbD approach.

• FMEA prioritized cryopreservation and fill-finish unit operations.
• DoE on cryoprotectant composition and freezing ramp rate.
• Parameter–Attribute Matrix (PAM) used.

Outcome.

• Viability range tightened to >75% with minimal batch discard.
• Cryopreservation and thaw SOPs revised.
• Fill volume and content uniformity optimized.

>75%

post-thaw viability range

10.5 Oligonucleotide Drug Product, Truncation Control

Background. A phosphorothioate oligonucleotide exhibited high levels of truncated species, affecting potency and triggering batch failures.

QTPP element. High-fidelity full-length sequence for antisense activity.

CQA identified. % full-length, % N-1 truncation, sequence integrity.

QbD approach.

• DoE on activator equivalents, detritylation duration, and coupling efficiency.
• Online monitoring with in-process capillary electrophoresis.

Outcome.

• Full-length purity improved from 89% to 95%.
• Truncation control translated into consistent biological activity.

89%

full-length purity (before)

95%

full-length purity (after)

Strategic value of QbD across the lifecycle

Organizations that integrate QbD early and comprehensively realize benefits that extend far beyond compliance, spanning regulatory flexibility, operational efficiency, cost control, and patient-centric quality (detailed in the Conclusion below). These advantages are particularly critical for modalities with tight process margins, short shelf lives, or complex CMC profiles, such as CGTs and vaccines.

11. Implementation Challenges and Mitigation Strategies

Despite its well-established regulatory support and proven value, implementation of QbD presents multidimensional challenges, technical, regulatory, operational, and cultural. These considerations must be anticipated and proactively addressed to realize the full benefits of QbD. The table below summarizes common hurdles and mitigation strategies encountered during QbD execution across small molecules, biologics, oligonucleotides, and advanced therapies (e.g., CGTs).

11.1 Summary of Key Challenges and Strategies

11.2 Modality-Specific Challenges

11.3 Strategic Enablers for QbD Success

To navigate these challenges, successful QbD organizations adopt a set of enabling practices:

• Platform Leverage: Use standardized process knowledge (e.g., cell culture platforms, purification trains) across programs to reduce development burden.
• Digital Infrastructure: Invest in integrated data lakes, analytics pipelines, and process historians to enable real-time decision-making.
• Regulatory Alignment: Early engagement with authorities (e.g., pre-IND, scientific advice meetings) ensures alignment on QbD expectations and mitigates late-phase surprises.
• Phased Implementation: Introduce QbD incrementally, focusing first on high-risk unit operations or CQAs with the greatest impact.
• Cross-Functional Governance: Establish a centralized CMC/QbD council to harmonize decisions, document rationales, and manage cross-silo communication.

12. Conclusion and Future Outlook

Quality by Design has redefined pharmaceutical development and manufacturing, transforming it from a historically reactive model into a proactive, science- and risk-based discipline. Anchored by global harmonization through ICH Q8–Q12, QbD embeds quality directly into the product and process by leveraging process understanding, predictive control, and lifecycle risk management.

What began as a framework for structured development is now a core element of regulatory expectation, operational excellence, and continuous improvement across modalities, from small molecule APIs to complex biologics, oligonucleotides, and cell and gene therapies (CGTs).

12.1 Strategic Value of QbD Across the Lifecycle

Organizations that integrate QbD early and comprehensively realize benefits that extend far beyond compliance:

These advantages are particularly critical for modalities with tight process margins, short shelf lives, or complex CMC profiles, such as CGTs and vaccines.

12.2 Evolving Trends in QbD

QbD continues to evolve with emerging scientific and regulatory developments:

• ICH Q13 (Continuous Manufacturing): Integrates QbD into end-to-end continuous platforms, demanding real-time data streams and PAT integration.
• ICH Q14 & USP <1220> (Analytical Procedure Lifecycle): Extend QbD principles into method design, validation, and maintenance.
• Digital QbD (dQbD): Leverages AI/ML, digital twins, and process modeling to simulate and optimize development with fewer physical runs.
• Regulatory Innovation Pathways: FDA’s Emerging Technology Program (ETT) and EMA’s Innovation Task Force actively encourage QbD-aligned technologies such as Multivariate Process Control (MVPC), model-based Real-Time Release Testing (RTRT), and digital batch release and CPV analytics.

12.3 Looking Ahead

In the near future, QbD will no longer be optional: it will be embedded in how the industry designs, develops, and controls medicines. As more national agencies align with ICH Q12 and FDA/EMA deepen their support for structured lifecycle management, the cost of non-compliance with QbD principles will increase. To prepare, leading organizations are:

• Building QbD centers of excellence (CoEs).
• Embedding QbD training across QA, RA, CMC, and operations teams.
• Investing in data infrastructure to enable predictive, connected quality systems.
• Using QbD principles to inform post-approval change strategy and regulatory interactions.

Final thoughts

Quality cannot be tested into products; it must be built in by design.Joseph M. Juran

As the pharmaceutical and biotechnology sectors face increasing complexity, from personalized therapies to globalized supply chains,QbD offers a sustainable, scalable, and patient-focused solution to ensuring product quality and performance. Adopting QbD is not just about meeting today’s regulatory expectations: it is about building the foundation for tomorrow’s innovation.