Transcript S10-2_Stewart Geary_Pharmacovigilance During Clinical

Pharmacovigilance During
Clinical Development
Stewart Geary, M.D.
USA
Eisai Co., Ltd. (Japan)
Pharmacovigilance During Clinical
Development
1
Outline
• The Challenge of Safety During Clinical
Development
• Causality Assessments During Clinical
Studies
• Expedited Reporting During Clinical
Development
• Risk Management Plans During Clinical
Development
• Data Safety Monitoring Committees
Pharmacovigilance During Clinical Development
2
The Challenge of Safety During
Clinical Development
We are still learning about how to improve
drug safety during clinical development
Pharmacovigilance During Clinical Development
3
Nonclinical Studies Do Not Predict All
Human Toxicity
H. Olson et. Al, “Concordance of the Toxicity of Pharmaceuticals in H
umans and in Animals”, Regulatory Toxicology and Pharmacology 32,
56–67 (2000)
4
Pharmacovigilance During Clinical Development
New Major Toxicities Discovered at Each
Stage of Clinical Development
H. Olson et. Al, “Concordance of the Toxicity of Pharmaceuticals in H
umans and in Animals”, Regulatory Toxicology and Pharmacology 32,
56–67 (2000)
Each Phase of Clinical Development is
likely to discover new safety issues
Phase I：61%
Phase II：29%
Phase III：10%
Pharmacovigilance During Clinical Development
5
Study Design Influences AE
Reporting
• More contact between Subjects and Investigators
increases event reporting
– Phase I trials performed with subjects admitted
to study center collect larger numbers of
nonserious events compared to later trials
• Frequency of lab tests can affect detected
incidence of lab abnormalities
• Both overall duration of trial and frequency of
study site visits can affect total numbers of
adverse events reported
Pharmacovigilance During Clinical Development
6
Study Design Influences AE
Reporting
• Common unrelated events during phase I trials
due to study site
– Headache
• (subjects going through caffeine withdrawal?)
– Diarrhea, constipation, abdominal discomfort
• (subjects adjusting to food at study site?)
– Sleepiness, difficulty sleeping, restlessness,
fatigue
• (psychosocial reactions to being in study site?)
Pharmacovigilance During Clinical Development
7
Placebo Lab Test Abnormalities
• Healthy volunteer studies with 14 days exposure
and subjects kept hospitalized/institutionalized
• 20.4% of the 93 subjects showed at least one ALT
value above the upper limit of the normal range
(ULN), and 7.5% had at least one value twice ULN
• Elevations occurred after the 1st week
– Due to imbalance in reduced physical activity
and increased calorie intake?
• Recovery was rapid after conclusion of study
(discharge from phase I facility)
Pharmacovigilance During Clinical Development
8
Br J Clin Pharmacol. 1999 July; 48(1): 19–23
Number of ALT elevations on placebo in healthy volunteer
phase I studies
Pharmacovigilance During Clinical Development
9
TeGenero TGN1412 Incident
• Pharmacodynamic activity to humanspecific targets is poorly predicted in
animal studies
• MABEL as an aid in setting initial dose for
first-in-human studies for biologic
compounds
• Some targets risk triggering toxicity
cascades
• Dose staggering for initial doses
Pharmacovigilance During Clinical Development
10
BIA 10-2474 Incident
• Even following the lessons of TGN1412
does not necessarily prevent catastrophic
outcomes
• Nonlinear pharmacokinetics?
• Step-function toxicity?
• Delayed toxicity in a setting of repeated
dosing?
Pharmacovigilance During Clinical Development
11
Causality Assessments During
Clinical Studies
ICSR Causality Assessments are
always uncertain and the safety
profile is determined by comparing
different treatment groups
Pharmacovigilance During Clinical Development
12
Principles of Individual Event Causality
Assessment
• Causality assessment is not an attempt to
“guess” treatment assignment
– Causality assessment by the reporter should be performed before
blind-break
– The assessment should not be changed after blind-break
• Unless there is new case information on the event or outcome
– Individual case SAE assessment for blinded reports should
assume subject received active drug ( just as assessment of
efficacy is made while blinded and assuming active treatment)
– When study is completed we do not go back and re-assess
individual case causality even if aggregate analysis suggests some
of the “not related” events are probably related to study drug
Pharmacovigilance During Clinical Development
13
Placebo Cases Often Considered
“Reactions” Prior to Unblinding
• Trial #1 1:1 Randomization active:placebo
– 10 SUSAR’s unblinded
– 5 were placebo, 5 were active
• Trial #2: 2:1 Randomization active:placebo
– 17 SUSAR’s unblinded
– 5 were placebo, 12 were active
• Trial #3: 2:1 Randomization active: placebo
– 19 SUSAR’s unblinded
– 5 were placebo, 14 were active
• A SUSAR on unblinding is almost as likely to be
assigned to placebo as active
– Better to know incidences of specific SAE’s in
placebo vs active
Pharmacovigilance During Clinical Development
14
CIOMS VI: Case Causality
Assessments
• Individual case causality assessment is subjective
• Comparison of incidence of “related” reactions
should be avoided
– Compare incidences of all events
• “It is the randomization that allows for causal
inference when comparing groups between
randomized arms, and consideration of the
investigator’s causality assessment is not helpful
for aggregate analysis.” (CIOMS VI)
Pharmacovigilance During Clinical Development
15
Causality Categories
• CIOMS VI recommended simple categories
“related” and “not related” with threshold
being a “reasonable possibility” of
relationship
– Different terms are interpreted differently
• Definitely, probably, possibly, unlikely, remote
– Studies comparing causality assessments of the same
reports by different observers show poor agreement
for various degrees of causality but better agreement
for overall decision of related or not related
– Terms such as “unlikely”, “remote” are often
misunderstood and should be avoided
Pharmacovigilance During Clinical Development
16
Causality Threshold in E2A and E2D
• Change in text regarding “cannot be ruled out”
E2A
E2D
Pharmacovigilance During Clinical Development
17
Expedited Reporting During
Clinical Development
Rules for expedited reporting
differ between ICH regions
Pharmacovigilance During Clinical
Development
18
What Cases Should Be Expedited
Reports?
• Clinical Development in ICH territories
– EU: SUSAR’s
– Japan: SUSAR’s
– US: Serious Unexpected Adverse Reactions
• BUT Regions Differ on defining “reactions” and
“unexpected”
– Expectedness: Reactions may be considered “expected”
earlier in Japan than in the EU and US
– “Reactions”: US FDA expects Sponsor to make this
decision according to “reasonable possibility” standard
regardless of Investigators’ opinion
Pharmacovigilance During Clinical Development
19
FDA IND Rule Effective March 28,
2011
• Changed reporting requirements to
improve quality of expedited reporting and
reduce unnecessary reports
– “Under former regulations, there may have
been over-reporting of serious adverse events
for which there was little reason to believe that
the drug had caused the event, complicating or
delaying FDA’s ability to detect a safety signal.”
Pharmacovigilance During Clinical Development
20
FDA IND Rule
• Clarified standard for considering a report
a “suspected adverse reaction”
– “any adverse event for which there is a
reasonable possibility that the drug caused the
adverse event. For the purposes of IND safety
reporting, ‘‘reasonable possibility’’ means there
is evidence to suggest a causal relationship
between the drug and the adverse event”
Pharmacovigilance During Clinical Development
21
“Evidence”?
• Examples of “evidence” may be:
– A single occurrence of an event that is uncommon and
known to be strongly associated with drug exposure
– One or more occurrences of an event that is not
commonly associated with drug exposure, but is
otherwise uncommon in the population exposed
– An aggregate analysis of specific events observed in a
clinical trial that indicates that those events occur more
frequently in the drug treatment group than in a
concurrent or historical control group
Pharmacovigilance During Clinical Development
22
FDA Proposed “Always Expedited”
•
•
•
•
•
•
•
•
•
•
Congenital anomalies,
Acute respiratory failure,
Ventricular fibrillation,
Torsades de pointe,
Malignant hypertension,
Seizure,
Agranulocytosis,
Aplastic anemia,
Toxic epidermal necrolysis,
Liver necrosis,
Acute liver failure,
Anaphylaxis,
Acute renal failure,
Sclerosing syndromes,
Pulmonary hypertension,
Pulmonary fibrosis,
Confirmed or suspected
transmission of an infectious
agent by a marketed drug or
biological product,
• Confirmed or suspected
endotoxin shock,
•
•
•
•
•
•
•
Pharmacovigilance During Clinical Development
23
Sponsor Decides
• Only the Sponsor’s causality assessment
counts in determining reportability to the
FDA
– “the sponsor is better positioned than the
investigator to assess the overall safety”
Pharmacovigilance During Clinical Development
24
Effect of Different Regulations
• FDA regulation does a better job of
focusing attention than reporting all
SUSAR’s
• For the same study, using the same
Investigator’s Brochure, the FDA will receive
far fewer Expedited Reports than EU or
Japan Regulators
– Applies also to reports received by
Investigators
Pharmacovigilance During Clinical Development
25
Risk Management Plans
During Clinical Development
The Development Safety Update
Report (DSUR) requires ongoing
determination
of
Identified
&
Potential Risks from the start of
clinical development.
26
Risk Management Plans
• Risk Management Plans (RMP) have
become a focus of post-marketing
pharmacovigilance
• Although regulations allow for request of
an RMP during clinical development, in
practice this rarely happens
• The Development Safety Update Report
(DSUR) requires ongoing designation of
Identified & Potential Important Risks
Pharmacovigilance During Clinical Development
27
DSUR
• DSUR or its equivalent required in all ICH
countries (EU/Japan/US)
• Section 19 requires summary of Risks
Pharmacovigilance During Clinical Development
28
DSUR
• The DSUR is required from the initiation of
phase I so this Risk summary is initially
based on limited, compound-specific
information
– Caution is necessary…
• Provides the core of a Development Risk
Management Plan
• “Important Missing Information”, which is
one area of Safety Concerns in the RMP is
not required in the DSUR
Pharmacovigilance During Clinical Development
29
Data Safety Monitoring
Committees
Pharmacovigilance During Clinical
Development
30
Data (Safety) Monitoring Committees
• Use started in 1960’s for clinical trials run
by US National Institutes of Health (NIH)
and Dept. of Veterans Affairs (VA)
• Independent committees which can review
unblinded data while a trial is ongoing for
efficacy or safety issues
• Use is now common in phase 2 and 3 for
mortality or major morbidity endpoints
Pharmacovigilance During Clinical Development
31
Independent Committees
• “A crucial lesson that should be learned from our
meta-analysis is that data on adverse events from
industry-sponsored randomised trials are
trustworthy only if an independent endpoints
committee is involved.”
– P. Juni et al, Lancet, (Correspondence) 2005,
365: 27.
– P. Juni et. al, “Risk of cardiovascular events and
rofecoxib: cumulative meta-analysis,” Lancet,
2004, 364: 2021-29.
Pharmacovigilance During Clinical Development
32
DMC Operation
• Charter lays out composition, data access and meeting
frequency
• Independent of trial Sponsor but often a Data
Management representative from the Sponsor is on the
committee to provide a liaison for data access
• Require expertise in therapeutic area, specific safety
issues and statistics
• Meetings have an open & closed portion
– And open & closed minutes
• DMC makes recommendations but Sponsor is
responsible for stopping study
Pharmacovigilance During Clinical Development
33
DMC Impact
• Supporting DMC requires extensive preparation and
definition of data that will be provided
–
–
–
–
Blinded, Grouped but blinded, or Unblinded?
SAE’s? AE’s? Lab Data? Study Endpoints?
Data provided pre- or post- QC?
Data only from this study or also from other studies and/or
postmarketing data?
• Performing an efficacy analysis can require adjusting
final analysis p-value needed for statistical significance
• DMC recommendations can result in sudden changes to
trial conduct
Pharmacovigilance During Clinical Development
34
SUMMARY
Pharmacovigilance During Clinical
Development
35
Summary
• Great care is needed during safety monitoring of clinical
trials because important new safety issues are discovered
at every stage of development and cannot be fully
predicted by nonclinical data
• Individual case adverse event causality assessment is
difficult, subjective and not as reliable as aggregate
analyses
• Reporting regulations during clinical development vary
considerably even between ICH countries
• The DSUR lays the groundwork for the Development
RMP
• Data Monitoring Committees play an important role in
ongoing safety
analyses
36
Pharmacovigilance During Clinical Development
Thank you for your kind
attention!
Pharmacovigilance During Clinical
Development
37