Good Pharmacovigilance Practices

1
Good Pharmacovigilance Practices

• Min Chen, RPh
• Associate Director,
• Division of Drug Risk Evaluation
• Office of Drug Safety, CDER

2
What is Pharmacovigilance?

• Is generally regarded as all postapproval
  scientific and data gathering activities relating
  to the detection, assessment, understanding, and
  prevention of adverse events or any other
  product-related problems
• This includes the use of pharmacoepidemiologic
  studies

3
Good Pharmacovigilance Practice starts by
acquiring complete data from spontaneous adverse
event reports
4
Characteristics of a Good
Case Report

• Adverse event(s) details
• Baseline patient characteristics
• Therapy details
• Time to onset of signs or symptoms
• Diagnosis of the event(s)
• Clinical course of the event and outcomes
• Laboratory data
• Any other relevant information

5
Good Medication Error Report

• Product(s) involved
• Sequence of events leading to the error
• Work environment in which the error occurred
• Types of personnel involved with the error
• Narratives- follow NCC MERP taxonomy (National
  Coordinating Council for Medication Error
  Reporting and Prevention)

6
Case Series Development

• Includes spontaneous case reports and scientific
  literature case reports
• Good database search strategies using MedDRA
  terminology
• Case definitions may be developed to provide
  consistent characterizations of the adverse
  events, and to facilitate retrieval of all
  clinically relevant cases

7
Assessing Causality

• It is rarely certain whether an event is product
  induced
• Features supportive of an association
• event occurred in the expected timeframe
• absence of symptoms prior to exposure
• absence of co-morbid conditions or use of
  concomitant medications
• () dechallenge, () rechallenge
• event consistent with the established mechanism
  of action of product
• Grouping case reports into probable,
  possible, and unlikely when appropriate

8
Safety Signals

• An apparent excess of adverse events associated
  with a products use
• a single well-documented case report may be
  viewed as a signal
• preclinical findings
• experience with other similar products in the
  class
• May be further assessed in terms of magnitude,
  population at risk, changes in risk over time,
  biological plausibility and other factors

9
Data Mining As a Signaling Tool

• Uses raw case report data and arrays of all
  drug-adverse events combinations
• Calculates expected (E) number of events for
  all drugs, then compares each drugs observed
  (O) to expected
• Various methods may be used to generate relative
  signal scores (intensity) of OE
• May be useful adjunct to routine safety signaling
  methods
• Further exploration and validation is needed

10
Safety Signals May Be...

• New unlabeled adverse events
• An observed increase in the severity or
  specificity of a labeled event
• An observed increase in the frequency of a
  labeled event
• New interactions
• Confusion with a products name, packaging or
  use, either actual or potential

11
Understanding Safety Signals

• Demographics - age, gender, race
• Effect of exposure duration and dose
• Relationship between concomitant medications and
  potential interactions and the risk of event
• Relationship between co-morbid conditions and the
  risk of event

12
Understanding Safety Signals (contd)

• Effects of lot-to-lot variation and differences
  in product formulation and the risk of the event
• Potential for an excess of adverse events given
  the disease being treated
• Estimates of the magnitude of risk or differences
  from known background rates

13
Questions for Public Comment

• How can the quality of spontaneously reported
  case reports be improved?

14
Questions for Public Comment

• What are possible advantages or disadvantages of
  applying data mining techniques (e.g., empirical
  Bayesian techniques, proportional reporting
  ratios) to spontaneous reports databases for the
  purpose of identifying safety signals?

15
Questions for Public Comment

• What are possible advantages or disadvantages
  of performing causality assessments at the
  individual case level?