Why We Set Specifications: Types of Quality Measures, Limits and Their Significance

1
Why We Set Specifications Types of Quality
Measures, Limits and Their Significance

• AAPS workshop on Specifications for
  Biotechnology and Biological Products
• October 6, 2004
• Barry Cherney, Ph.D. Deputy Director
• DTP/OBP/CDER, FDA

2
Overview

• Define terminology to be used for this workshop
• Describe the types of Quality Measures and their
  use
• Describe the types of limits
• Describe general principles for setting
  acceptance criteria

3
General Goal

• To ensure products are made with the appropriate
  identity, strength, quality, purity, and potency
  as they may relate to the safety or effectiveness
  of the product.

4
Comprehensive Quality Control Strategy

• Process Product Testing
• - Control of Raw Materials
• Process Validation - Characterization
• In-Process Controls - Release Testing
• In-Process Testing
• cGMPs (QU) - Stability Testing

5
Utility of Quality Measures

• Indicates freedom from adventitious agents and
  other contaminates
• Indicates critical product quality attributes are
  representative of attributes found in clinical
  trial material
• Indicates consistency of the manufacturing
• Monitors process
• Controls product heterogeneity
• May act as a surrogate marker to ensure product
  quality is not impacted. Example Met Oxidation

6
How Does One Measure Product Quality?

• A test
• A specific analytical procedure including
• Sampling plan
• Acceptance criteria for determining appropriate
  results
• Reporting the final result (Reportable value
  individual results or mean)
• These elements are components of both
  Specifications and Limits (Action)

7
Types of Quality Measures

• Specifications
• Established at important process steps for
  important parameters
• Establish the set of criteria to which a material
  should conform to be considered acceptable for
  its intended use (Q6B)
• Used to confirm product quality (Q6B)
• Represents a subset of analytical methods used
  to fully characterize the product

8
Types of Quality Measures

• Limit Tests
• Established for less critical parameters
• Typically established as an in-process test
• May not be part of the characterization testing
• Establishes acceptable test results beyond which
  an investigation should be conducted
• Characterization Tests

9
Regulatory Implications

• Exceed Specification
• Out of Specification Investigation
• If OOS is confirmed, material should be rejected
  211.165(f)
• May submit a PAS for product exception/change in
  acceptance criteria or implement a reprocessing
  protocol but must be approved by the Agency
• Exceed Limit
• Deviation report and investigation is conducted
• Outcome determines acceptability for further
  manufacture or release
• Review by the Agency is not required for release

10
Working Definitions

• Specification an acceptance criterion that
  product must meet and is registered with
  regulatory agencies
• Limit an acceptance criterion that product
  should meet and is registered with regulatory
  agencies
• Acceptance Criterion a numerical value range or
  other suitable measure that product quality is
  measured against
• Release Specification - an acceptance criterion
  that the product must meet at release
• Release Limits an acceptance criterion that the
  product should meet at release
• Shelf-life Specification - an acceptance
  criterion that the product must meet throughout
  its shelf-life

11
Where Are Specifications Established

• Raw Materials/ Excipients/Container/closure
• In-process materials/intermediates (e.g.
  mycoplasma test/ PEG)
• Drug Substance release
• Drug Product release
• Stability

12
Where Do We Use Limits?

• In-process
• Examples
• Bioburden
• Endotoxin
• Step yields
• Specific activity

13
What Could Limits be Established For?

• Drug Substance and/or Drug Product Release
• For monitoring product quality (e.g. Bioburden)
• To capture out of tend results (out of process
  capability limits)
• For consistency measures
• To ensure product quality over the shelf-life?
• Stability
• For consistency measures

14
What Conditions Should be Met to Replace Specs
With Limits

• Additional downstream steps that impact outcome
• Demonstration that an quality attribute does not
  impact S and E?
• Detailed procedures to describe actions when you
  fail to meet a specific release limit?
• An effective and scientifically rigorous Quality
  Unit?

15
For What Tests Could Limits be Applied?

• Appearance
• Identity (Desired product)
• Purity (freedom from extraneous material, product
  variants)
• Impurities (Product and process related
  impunities)
• Potency
• Quantity
• Other quality characteristics (pH, sterility
  endotoxin, moisture, and particulates)
• Safety from Adventitious Agents

16
Setting Acceptance Criteria

• Based on data obtained from
• Lots used in clinical studies (gold standard)
• Lots used in non-clinical studies
• Manufacturing history (e.g. process capability)
• Stability studies
• Relevant developmental data
• Analytical methods

17
Non Clinical Clinical Experience

• Clinical material (the gold standard)
• Quality of the product made at commercial scale
  should be representative of lots used in the C/NC
  studies (Q6B)
• Establishes a direct link between a quality
  attribute and safety and efficacy
• Beneficial to fully characterize and use
  multiple lots in clinical studies
• Non-clinical data
• Animal models can be useful in assessing the
  impact of specific quality attributes have on
  safety and efficacy
• Rationale for linking a quality attribute to S
  or E

18
Non Clinical Clinical Experience

• Clinical/non clinical data can be mined to help
  establish specifications, if the studies are well
  thought out regarding
• The specific quality attribute and observed
  variation
• Process capacity of the manufacturing process
• Ability to track specific lots in clinical
  studies
• Consider the quality attributes when choosing
  lots for clinical/ non-clinical studies

19
Manufacturing History

• Acceptance criteria (for specs and limits) should
  reflect the manufacturing history (Q6B)
• Terminology
• Process capability - the ability of a
  controlled and stable manufacturing process to
  produce a product with defined quality
  attributes
• Process capability limit The range (extreme)
  of measurements for a quality attribute that
  represents the typical variation observed with a
  controlled and stable manufacturing process
• Quality Attribute - a measured product
  characteristic that is selected for its ability
  to help indicate the quality of the product

20
Manufacturing History

• Often the major element in setting specs
• Amendable to statistical analysis (if adequate
  sample size)
• Mean 3 standard deviations the default range
  for setting release specs
• Allows for ranges in quality attributes that are
  greater than those observed in the clinical
  material and beyond process capability limits

21
Model of Limits
22
Manufacturing History

• Should all available data be incorporated into a
  statistical analysis for estimating process
  capability?
• Lots that were evaluated with imprecise assays
• Early process that was not tightly controlled or
  lots from a different process
• Lots that had significant manufacturing
  deviations
• Careful consideration should be given to the data
  set included in a statistical analysis
• Limited utility if few lots have been produced

23
Developmental Data

• In depth characterization of the desired product
  (and its variants) can help establish links
  between QA and S E

24
Stability

• Specifications should be established to ensure
  continued purity, potency and safety throughout
  the dating period
• Stability tests should monitor quality attributes
  that potentially change over time
• Acceptance criteria is not based simply on the
  results of stability testing (including stress
  testing and shipping and handling studies) but on
  links to clinical and non clinical data
• How do we establish these links?

25
Lessons Learned From Comparability Studies

• Differences in quality attributes that are within
  spec but outside process capability limits have
  sometimes had a significant consequence
• The more you know how a quality attribute
  impacts on safety and efficacy the less you may
  need to do in assessing differences between pre
  and post change products
• Your understanding of quality attributes could
  impact on the level of control required for each
  attribute

26
Acknowledgements

• Division of Therapeutic Proteins
• Christopher Joneckis
• Andrew Chang
• Philip Krause
• Steven Kozlowski
• Anthony Mire-Sluis

27
Breakout Session AWhy We Set Specifications

• What are the goals of specifications, release
  limits, and process capability limits?
• What should be the relationship of
  specifications, release limits, and process
  capability limits?
• What actions are warranted when one or the other
  of these limits are violated?