Risk-Based Sourcing for Bovine Materials in FDA-regulated Medical Products: Introduction and General Considerations FDA TSE Advisory Committee Meeting Holiday Inn Silver Spring MD 12 February 2004 David M. Asher, MD Laboratory of Bacterial, Parasitic and Unconventional Agents Division of Emerging & Transfusion-Transmitted Diseases Office of Blood Research and Review Center for Biologics Evaluation and Research United States Food and Drug Administration e-mail address: [email protected]
20
Embed
Risk-Based Sourcing for Bovine Materials in FDA-regulated Medical Products: Introduction and General Considerations FDA TSE Advisory Committee Meeting.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Risk-Based Sourcing for Bovine Materials in FDA-regulated Medical Products:
Introduction and General Considerations
FDA TSE Advisory Committee MeetingHoliday Inn
Silver Spring MD
12 February 2004
David M. Asher, MDLaboratory of Bacterial, Parasitic and Unconventional Agents
Division of Emerging & Transfusion-Transmitted Diseases
Risk-Based Sourcing for Bovine Materials in FDA-regulated Medical Products:
Introduction and General Considerations
Recent TSE-related events with implications for safety of FDA-regulated products Presumptive transfusion-transmitted vCJD Cow with BSE in Washington State
Human blood/tissues and related products Products containing or manufactured with
bovine materials Risk analysis and TSEs Existing safeguards (“firewalls”), enhancements Possible additional safeguards for injectable
biologics and similar products: discussion points
Risk Analysis for Products:Conventional Definition
Risk Assessment (Paradigm of Natl Acad Sci USA /Natl Res Council report 1983)
1. Identify hazard
2. Characterize hazard (dose-response)
3. Estimate probable exposure to hazard
4. Characterize risk (from estimated probable exposure and dose-response)
Risk Management (based on assessment)
Risk Communication
Risk Assessment for Products:General Elements
of Microbiological Risk
Source of raw materials (contamination)
Manufacturing process Reduction of contamination
(ability to inactivate or remove contaminant)
(? Concentration/increase of contaminating agent)
End-use (frequency, amounts, routes)
“Effective” Exposure to TSE Agent (Kimberlin):
An Exposure Sufficient to Infect a Recipient
Dose of agent (infectivity/volume x total volume) Route of exposure
Expresses risk as an overall probability Tries to represent the complexity of real
situations as “scenarios” (called System or Fault Tree or Failure Modes-and-Effects Analysis: multiply individual probabilities of a series of system failures leading to an adverse event)
Requires collection of adequate, accurate, quantitative data to be reliable
Allows for variability and uncertainty (best expresses probabilities as distributions, not single-point estimates)
Identifies variables with the greatest impact on the risk estimates (Sensitivity Analysis)
Quantitative Risk Assessment: Dealing with Variability and
Uncertainty(EC DG SANCO)
Describe all assumptions and constraints.
Represent parameters as probability distributions rather than as single-point-estimate values.
Perform Sensitivity Analysis to demonstrate the influence of assumptions used in the risk assessment on the final risk estimate.
(Answers question: “What if we are wrong about an assumption?”)
Advantages and Weaknesses of Quantitative Risk Assessment
Transparency Assumptions and methods clearly articulated Failure scenarios available for independent
review: elements of risk less likely to be overlooked
Requires collection of adequate, accurate, quantitative data to be reliable For unprecedented hazards no actuarial
information available to estimate probabilities Absent precedents, probabilities for each failure
step are estimated at least partially subjectively Expressing subjective probabilities
numerically implies unjustified level of confidence
Assumes no surprises
PhRMA BSE Risk Model (Bader & al. 1998) and Uncertainty
Bader F et al. Assessment of risk of bovine spongiform encephalopathy in pharmaceutical products. Biopharm 1998;11(1):20-31 and 11(3):18-30
Probability Sources of Uncertainty
Cow infected (herd, age) Quality of surveillance, inspection etc.
Batch contaminated (tissue, slaughter technique)
High-infectivity tissue cross contamination, etc.
Human infectivity per tissue Species barrier estimate
Human infectivity per batch after processing
Limited process validation data, questionable relevance
Human infection per dose Uniformity of susceptibility
Human infections per course Uniformity of susceptibility
Problems with All Risk Assessments
Experts may disagree about assumptions. Elements of risk Probabilities
Models assume no surprises. Bias is unavoidable.
General bias Non-independent probability estimates (Experts may be optimists or pessimists.) Bad-news bias (Probability of a dramatic [recent]
adverse event happening again is overestimated.) Others
Expert bias Contractor bias (“piper’s” bias): seek a desired
outcome Manager bias (shop the experts): seek a desired