Sei sulla pagina 1di 32

Implementation of Quality by

Design:
An Office of Biotechnology
Products Perspective

Steven Kozlowski, M.D., Director


Office of Biotechnology Products OPS/CDER

ACPS 10/5/06
Overview
• OBP Products & Quality by Design
• Relevant Product Attributes
• Manufacturing Process
• Implementation Plans
Office of Biotechnology Products
• Therapeutic Proteins
– Growth Factors
– Enzymes
– Cytokines
– Chemokines
– Angiogenic factors
– Toxins
– Soluble Receptors/Receptor antagonists
• Monoclonal antibodies
–Related Products
Office of Biotechnology Products
• These proteins are usually produced from:
–Cell culture
• Recombinant & non-recombinant
• Mammalian, bacterial, yeast, etc.
–Transgenic plants & animals
• Products transferred from CBER to CDER in
October 2003
• ONDQA regulates some protein products
Structure Fab = ~1/3 mAb
of complex
molecules
1° structure
Statin

• higher order structure
• post-translational
modifications
• heterogeneity o
11 A

o
87 A
Statin MW ~400 Da
Monoclonal Ab MW ~150,000 Da
PDB 2IG2, 1HW8
Relationships
Safety Manufact.
Efficacy Process
(SE) (P)

Quality
Attributes
(A)
Comprehensive QbC Strategy
Process Product

• Facilities and Equipment • Method Validation


• Control of Raw Materials • Release Testing
• In-Process Testing (PAT) • Characterization
• In-Process Controls • Stability Testing
• Process Robustness (FED)
• Process Validation
• cGMPs (QC/QA)
What is “Quality by Design”?
• Quality
– “Good pharmaceutical quality represents an acceptably low
risk of failing to achieve the desired clinical attributes.”
• Quality by Design (QbD)
– “Means that product and process performance
characteristics are scientifically designed to meet specific
objectives, not merely empirically derived from performance
of test batches.”

Janet Woodcock (2004)


SE P

Quality
by
Design

SE P

A Moheb Nasr
Overview
• OBP Products & Quality by Design
• Relevant Product Attributes
• Manufacturing Process
• Implementation Plans
How Much of the Iceberg
(Quality Attributes) Can We See?

• Release tests

• Characterization

? • Process
How Much of the Iceberg
(Quality Attributes) Can We See?

• Release tests

• Characterization
•MS, NMR
•Orthogonal methods

• Process
?
Attributes & Combinatorics
pyro-E pyro-E
O O • Pyro-Glu (2)
D D
D D • Deamidation (3 x 2)
O O
• Methionine
G G
oxidation (2 x 2)
D D
• Glycation (2 x 2)
• High mannose,
G G
G0, G1, G1, G2 (5)
K K • Sialylation (5)
• C-term Lys (2)

• (9600)2≈ 108 • 2 x 6 x 4 x 4 x 5 x 5 x 2 = 9600


Relevant Attributes SE P

• … those molecular and biological A


characteristics found to be useful in
ensuring the safety and efficacy of the
product (Q6B)
• Can these attributes be defined?
– Often difficult
– Default is to look at many attributes
Biological Characterization
Structure/function



Biological Activity

Purified/Induced Variants
Matrix

Clinical Lot Extremes

Developmental Lots
One to some lots

Stressed Lots
Clinical Lots
Many lots

Multiple Binding/Cellular Assays

Small Animal/Complex Bioassay

Clinical Pharmacology (PK/PD)

Clinical Studies

Validated Bioassay
CQA Multivariate Range

Charge
SE P

2
rm
A

fo
co
Glyc
o

Gly
form
1
The Desired Product
• Desired attributes of the API (Focus for
Biotech)
─ Opportunity for rational protein engineering
• Customize quality
─ Increase attributes that are desirable
─ Limit attributes that negatively impact
product quality (via process or product)
• Understanding structure/function
relationships is critical for this approach
Protein Engineering
• Reduce tendency for aggregation
─ Block free sulfhydryl groups
─ Alter amino acid sequence based on
computational predictions
 Glycine repeats or proline insertions
 Maintenance of net charge
 Alternate polar and non polar residues
 Clusters of hydrophobic residues
−Human Calcitonin (Zurdo et al PNAS 2005)
−Immunogenicity considerations
QSE…by Design
• Safety & Efficacy by Design
• Improving function/new properties

• Increasing Bioavailability
• Reducing immunogenicity
– Selective technologies in development such as
phage, ribosome and yeast display
• Quality by design
– These strategies can also be used to select for
product quality and manufacturability
Overview
• OBP Products & Quality by Design
• Relevant Product Attributes
• Manufacturing Process
• Implementation Plans
What is the Relationship SE P

Between Attributes & Process? A

• Release tests

• Characterization
•MS, NMR
•Orthogonal methods

• Process
?
What is the Relationship SE P

Between Attributes & Process? A

• Release tests

• Characterization
•MS, NMR
•Orthogonal methods

• Process
?
SE P Translating CQA to a SE P
Manufacturing Design
A A
Space
Charge

Time
2
m

dia
for
co

Me
Glyc Agit
of
Gly

orm ation
1
Critical Product Critical Process
Attributes Parameters
The Desired Process
• Desired attributes of the API or drug product
– Opportunity for rational process design
• Choose unit operations that select for
– select for desired product attributes
– minimize impurities/contaminants
– are robust
• Order unit operations to maximize efficiency
– minimize buffer changes
– consider impurities introduced by each step
• Process control
– consider impact of variable inputs
– parameters set based on maximizing multiple variables
– for critical steps- ideally real time & based on a knowledge base
(PAT)
Examples of Problematic Process
Designs
• Processes that promote product variability
• Heat treatment step added after aggregate
clearance step
• Process performed at room temperature with
negative impact on quality
• Recloning is used to establish new WCB
• Processes that are difficult to control
• Roller bottle processes (open, multiple
fermentations difficult to control)
• Poor formulation (stability & interactions)
Overview
• OBP Products & Quality by Design
• Relevant Product Attributes
• Manufacturing Process
• Implementation Plans
Implementation of QbD
• Benefit from other OPS knowledge on implementation
– e.g. ONDQA pilot
• OBP participation in agency CRADAs to facilitate QbD
• OBP Structure
• Mixture of research/reviewers and full time reviewers
• Experience with new technologies, fermentation/purification
• Expertise in biological characterization

• Encourage industry to engineer proteins for quality as well as safety and efficacy
– “desired product”
• Encourage industry to consider “desired process”
Small Steps
• Product Testing
– If no likely impact on S and E don’t include as a specification
(no rejection limit)
 use as a process consistency measure, where exceeding a limit
initiates an investigation
 if not a consistency measure, drop the test entirely
– Need to discuss more extensive in-house reviewer training
• Process Changes
– Strategy to assess risk
– Select classes of change based on
• supplement load
• variable risk to product
– FDA/industry forums targeted to create risk map for a single
class of change
– Publication as white papers
Small Steps (cont.)
• Complex API pilot
– Encourage manufacturers to generate and submit data on
characterization of structural attributes
• the use of multiple biological assays for characterization
 the generation of a product attribute space
 consideration of associated relief (e.g. greater flexibility for
comparability protocols)
– Need to consider & discuss further
• Platform Strategies
– monoclonal antibodies may share a great deal of structure
– efficient use of prior knowledge
– long history of a regulatory path for modular/generic validation
(Mab PTC)
• underutilized
– currently of interest to many biotech companies
– conference to put interested parties together
Question to ACPS
• Should FDA develop a pilot program and/or
initiative to explore specific QbD issues that
are important for biotechnology products?
• Such a program could focus on issues such
as:
– (1) general approaches to CQA for complex API
– (2) platform approaches/extrapolation for highly
related complex APIs.
Quality Life Cycle Definitions?
Design -Evaluation Identify
Propose (Critical/ Key
(Desired Attributes/Process) Parameters/ &
Ranges)
Development
Knowledge Base Studies (DOE)
Optimize Product-specific
(Continuous Process-specific
Improvement
Innovation) Quality Risks

Confirm
(Control/ Predict)
Monitor Conformance Lots
Post-Approval Commercial “Validation Studies”
Production
Credits
• Barry Cherney
• Patrick Swann
• Moheb Nasr
• Keith Webber
• Ajaz Hussain