Risk Management for Complex Pharmaceuticals:

1
Risk Management for Complex Pharmaceuticals
An Awareness Topic

• Steven Kozlowski, M.D., Director
• Office of Biotechnology Products OPS/CDER

ACPS 10/6/06
2
Overview

• Background
• Complex Product Risk Assessment
• Less than the Whole 9 Yards
• Future Considerations

3
Risk Management Tools (ICH Q9)

• Basic risk management facilitation methods
  (flowcharts, check sheets etc.)
• Failure Mode Effects Analysis (FMEA)
• Failure Mode, Effects and Criticality Analysis
  (FMECA)
• Fault Tree Analysis (FTA)
• Hazard Analysis and Critical Control Points
  (HACCP)
• Hazard Operability Analysis (HAZOP)
• Preliminary Hazard Analysis (PHA)
• Risk ranking and filtering
• Supporting statistical tools.

4
Risk Management (ICH Q9)

• Principles
• Link to patient protection
• Effort commiserate with potential risk

5
Risk Assessment (ICH Q9)

• Risk Probability X Severity
• What might go wrong?
• What is the likelihood (probability) it will go
  wrong?
• What are the consequences (severity)?

• Use in Quality Management Regulatory Operations
• Inspections Audits
• Facilities, Equipment Utilities
• Materials Management
• Manufacturing Change Control
• Assessment of Product Quality
• Complex Products

6
FMEA Bottom Up Semi-Quantitative
7
FTA Top Down Quantitative
8
Quantitative Risk Assessment
Probability
Major
Important
Catastrophic
Minor
Severity
9
Probabilistic Risk Assessment (PRA)

• What it means?
• Michael G. Stamatelatos, NASA
• Includes uncertainty of quantitation
• Modeling for unknown information
• Integration many initiating events
• Examples of modeling
• Fayssal M. Safie, Chad Hoffman, Rich
  PughAIAA Pub
• Similarity Method
• Probabilistic Structural Analysis
• Distributions of important variables
• Simulations (Monte Carlo, etc.)

10
NASA PRA
Fayssal M. Safie, Ph.D., CRE
11
Overview

• Background
• Complex Product Risk Assessment
• Less than the Whole 9 Yards
• Future Considerations

12
Structure of complex molecules

• 1? structure
• higher order structure
• post-translational modifications
• heterogeneity

Monoclonal Ab MW 150,000 Da
13
Attributes Combinatorics

• 2 x 6 x 4 x 4 x 5 x 5 x 2 9600

• (9600)2 108

14
Laurence J. Peter, American business humorist

• Some problems are so complex
• that you have to be highly intelligent
• and well informed
•  

just to be undecided about them.
15
Probabilistic Risk Assessment

• People tend to think that lack of data is a
  reason not to perform a PRA, but the exact
  opposite is true
• PRA is generally used for low-probability and
  high-consequence events for which insufficient
  statistical data exist.
• If enough statistical data exist to quantify
  system or sub-system failure probabilities, use
  of some of the PRA tools may not be necessary
• Michael G. Stamatelatos, NASA
• May serve as organizing framework
• Bilal Ayyub, U of MD

16
Protein LogicDiagram
Protein
17
Functional Event Sequence Diagram
Oxidation Site 1
18
Safety
Is Oxidation Without Impact On
Safety? (Immunogenicity)
Acceptable Product
Oxidation Site 1
Y
Y
Is Level Low Enough (or High Enough) not to
Generate an Immune Response?
Batch Failure
Does the Immune Response Lack Clinical
Significance?
Detectability of Oxidation with Impact?
Clinical Failure
19
Functional Event Sequence Diagram
Is Oxidation Without Impact on Activity?
Y
Oxidation Site 1
Acceptable Product
Is Level Low Enough not to Matter?
Y
Batch Failure
Detectability of Oxidation with Impact?
Clinical Failure
20
Event Tree (Quantification)
End State
Oxidation Site 1
Low Level
No Impact
Scenario
Detectable
21
Assigning Probabilities

• Actual Data
• Similarity Method
• Same Product
• Models (animals, bioassays)
• NASA Ground tests
• Related Product or Component
• Direct Data (clinical)
• Models (animals, bioassays)
• Probabilistic Structure Analysis
• e.g. probability of docking to all known
  receptors
• Far off

22
Biological Activity Matrix
23
Assessing Relatedness for Assigning Probabilities

• Same primary sequence
• Alignment in shared framework/domains
• BLAST surrounding sequences
• T-cell-like
• Protein Structure Databases
• B-cell-like
• Molecular Descriptors (Cramer)
• In silico immunogenicity prediction
• Need measures of uncertainty associated with each
  strategy used

24
Evolution of Monoclonal Antibodies
25
Monoclonal Antibodies as a Risk Management
Platform

• 50 of licensed MAbs are IgG1
• It is estimated that greater than 50 of Mabs
  under development are humanized or human IgG1s
• 50 of products may share 95 of their sequence
• Valuable human in vivo MAb data
• Isoform based PK data
• Large safety database for shared attributes
• All IgG1 products
• Endogenous IgG1 variants
• Radiolabeled antibodies imaging (generally not
  humanized)

26
Monoclonal Antibodies as a Risk Management
Platform

• Isoform bioassays
• Fc binding Effector Function
• Nature of target interaction
• Soluble, host cell based or on pathogen
• Accessibility of target
• Ability of target to signal
• Role of effector function
• Cross-reactivity

27
Overview

• Background
• Complex Product Risk Assessment
• Less than the Whole 9 Yards
• Future Considerations

28
Even Without a PRA

• Nothing is more difficult, and therefore more
  precious, than to be able to decide.
• Napoleon Bonaparte
• Even a correct decision is wrong when it was
  taken too late.
• Lee Iacocca

29
FMEA for Product Quality
Patrick Swann
30
Combination Products

• Primary Mode of Action
• Driver of jurisdiction between FDA Centers
• Toxic component overrides targeting
• Administrative Ease vs Risk Assessment
• Risk Assessment more Appropriate for Resource
  Allocation

31
Antibody Conjugate
P? Probabilistic Risk Assessment
32
Very Complex Combinations
33
Overview

• Background
• Complex Product Risk Assessment
• Less than the Whole 9 Yards
• Future Considerations

34
Considerations

• Follow up on prior programs
• Draft Report on Transdermal Patches
• Plan for Conference on Risk Management
  Pharmaceuticals
• Bilal Ayyub, U of MD
• Further OPS education on Risk Management
• Seminar Series
• Consider approached to encourage exploration of
  of risk management strategies for complex
  products
• If Antibody Platform approach used in QbD
• Strategize to best utilize Risk Management
• Best initial system for PBR

35
Homework

• Fault Tree Analysis for above product
• Assign probabilities to all possibilities
• Due 5PM today
• Alsoconsider relative importance of investing in
  risk management for complex products