Concept
Paper:
Premarketing Risk Assessment
March 3, 2003
PDF
version of this document
DRAFT
This concept paper is being
distributed for comment purposes only.
Submit comments to the Dockets Management
Branch (HFA-305), Food and Drug Administration, 5630 Fishers Lane, rm.
1061, Rockville, MD 20852; e-mail
fdadockets@oc.fda.gov;
or on the internet at
http://www.accessdata.fda.gov/scripts/oc/dockets/edockethome.cfm.
Written or electronic comments on the concept papers will be accepted
until April 30, 2003.
All comments should be identified with
docket number 02N-0528 and wherever possible, reference the comments to
specific sections or line numbers of the document.
For questions on the content of this draft
document contact Barbara Gould, 301-827-2504.
If you plan to submit comments on this
concept paper, to expedite FDA review of your comments, please:
- Clearly explain each issue/concern and,
when appropriate, include an alternative proposal and the rationale
and/or justification for employing the alternative.
- Identify specific comments by line
numbers; use the PDF version of the
document whenever possible.
I. INTRODUCTION
In accordance with Section VIII of the
PDUFA III Reauthorization Performance Goals and Procedures, the CDER/CBER
Risk Assessment Working Group is drafting a guidance for industry on good
risk assessment practices during drug and biological product1
development. This concept paper is intended to facilitate public
discussion on the content of the draft guidance by outlining FDA's
proposed approach and requesting comment. Specifically, this concept paper
presents FDA's preliminary thoughts on:
- Important risk assessment concepts
- Generation and acquisition of safety
data during product development
- Analysis and presentation of safety
data in an application for approval
II. IMPORTANT RISK ASSESSMENT CONCEPTS
Risk assessment is the process of
identifying, estimating, and evaluating the nature and severity of risks
associated with a product. Risk assessment occurs throughout a product's
lifecycle. To develop a risk management plan and perform pharmacovigilance
after approval, it is important to have as good an idea as possible of the
product's underlying risks and benefits prior to approval. This process
entails ensuring that the body of evidence generated by the clinical
trials not only defines the product's effectiveness, but also
comprehensively describes its safety (as required by the Food, Drug and
Cosmetic Act, which calls for the conduct of all tests reasonably
applicable to evaluate a drug's safety).
This concept paper focuses on risk
assessment during clinical development, particularly in phase 3 studies.
It does not discuss preclinical safety assessments (i.e., animal toxicity
testing) or clinical pharmacology programs, because these issues are
addressed sufficiently in current FDA and International Conference on
Harmonization (ICH) guidances.2
However, we emphasize that good clinical risk assessment depends on the
performance of comprehensive preclinical safety assessments and a
rigorous, thoughtful clinical pharmacology program (including elucidation
of metabolic pathways, drug-drug interactions, and effects of hepatic
and/or renal impairment).
No, this concept paper focuses solely on
risk assessment based on safety data generated during product development.
Risk assessment based on data generated from observational data sources
(including case reports, case series and pharmacoepidemiologic studies)
obtained after a product is marketed is addressed by a separate concept
paper entitled Risk Assessment of Observational Data: Good
Pharmacovigilance Practices and Pharmacoepidemiologic Assessment. Programs
intended to manage risk could stem from either premarketing or
postmarketing risk assessment efforts or both. Such programs are addressed
by a third concept paper entitled Risk Management Programs.
III. IMPORTANT CONSIDERATIONS IN
GENERATING RISK INFORMATION DURING CLINICAL TRIALS
The design of a product's clinical trials
program is critical in ensuring that sufficient safety data are generated
to allow for approval of the product, as well as to provide data to allow
for proper risk management and to inform post-marketing safety assessment.
Since many aspects of clinical development have previously been addressed
in FDA and ICH guidances,3 this concept
paper presents FDA's thoughts on selected issues as they apply to optimal
risk assessment.
The ideal size of a safety database
supporting a new product depends on a number of factors, including the
novelty of the product, the intended population, the proposed indication
(e.g., a treatment for a life threatening disease vs. a symptomatic
treatment) and the intended duration of use. In addition, safety concerns
identified in the preclinical safety assessment, or safety signals seen in
early clinical or human pharmacology studies, could suggest that more
intensive safety assessments (including greater patient exposure) would be
appropriate.
No guidance currently exists on
determining the appropriate size of clinical safety databases for products
intended only for acute use or for serious and life-threatening diseases,
although 21 CFR part 314 (subpart E) suggests that approval may occur
after phase 2 if benefit is established (i.e., without specific large
scale safety studies). On the other hand, most trials designed to show a
mortality advantage would be large in the first place and, if successful,
would often by themselves demonstrate an acceptable balance of benefit to
risk.
FDA would be interested in input on
what general guidance could be provided on appropriate sizes of databases
for products intended only for acute use and/or for serious and
life-threatening conditions. FDA is also interested in input on the
proposals below, related to safety assessments of chronically administered
drugs for non-life threatening conditions.
For products intended for long-term
treatment (e.g., chronic or recurrent intermittent) of
non-life-threatening conditions, the ICH has recommended that 1500
patients be exposed to the investigational product.4
However, the ICH guidance does not specify what patients should be counted
towards the 1500 patient target. For chronic use products that are novel
in mechanism or class, we believe the 1500 patients should include only
those who have been exposed to the product in multiple dose studies of
four or more weeks' duration, as many adverse events of concern (e.g.,
hepatotoxicity, hematologic events) do not usually appear with shorter
exposure. Also, ideally, the 1500 patients should have been exposed to
doses equal to or exceeding the lowest proposed dose, with a substantial
representation of patients exposed at or above the highest proposed doses.
In addition, the ICH guidance recommends that 300 to 600 patients be
exposed for 6 months or more, with at least 100 being exposed for 12
months.
The ICH E-1 guidance provides a number of
considerations that would suggest the need for a larger database,
including:
1. When "there is concern that the
drug would cause late developing adverse events, or cause adverse events
that increase in severity or frequency over time. The concern could
arise from:
- Data from animal studies;
- Clinical information from other
agents with related chemical structures or from a related
pharmacologic class;
- Pharmacokinetic or pharmacodynamic
properties known to be associated with such adverse events."
2. When "there is a need to
quantitate [sic] the occurrence rate of an expected specific
low-frequency adverse event. Examples would include situations where a
specific serious adverse event has been identified in similar [products]
or where a serious event that could represent an alert event is observed
in early clinical trials."
3. When "needed to make
risk/benefit decisions in situations where the benefit from the product
is either (1) small (e.g., symptomatic improvement in less serious
medical conditions) or (2) will be experienced by only a fraction of the
treated patients (e.g., certain preventive therapies administered to
healthy populations) or (3) is of uncertain magnitude (e.g., efficacy
determination on a surrogate endpoint)."
4. When "there is concern that a
[product] may add to an already significant background rate of morbidity
or mortality, and clinical trials need to be designed with a sufficient
number of patients to provide adequate statistical power to detect
pre-specified increases over the baseline morbidity or mortality."
In addition to the considerations provided
in the ICH guidance, other reasons why a larger database could be
appropriate include:
1. The proposed treatment is for a
healthy population (e.g., chemoprevention, preventive vaccines)
2. A very safe alternative to the
investigational product is already available
3. There is the potential for rapid
exposure to a large population.
B. What are some characteristics of
an ideal safety database?
The composition of an appropriate safety
database for a new product would be determined on a case-by-case basis.
Ideally, however, all programs would include:
Currently, it is common in many clinical
programs for much of the patient exposure and almost all of long-term
exposure to come from single-arm or uncontrolled studies. In most cases,
it would be preferable to have controlled safety data, including long-term
safety data, to allow for comparisons of event rates and for accurate
attribution of adverse events. Control groups could be given a placebo or
an active product, depending on the disease being treated. The usefulness
of comparators in longer-term safety studies (versus uncontrolled safety
trials) depends on factors such as the background rates of the adverse
events of interest. Generally, events that occur rarely and spontaneously
(e.g., idiopathic hepatitis) do not need a control group to be
interpreted. On the other hand, control groups are essential for detecting
changes in rates of events that occur frequently in the population (e.g.,
death in patients with Alzheimer's). This is particularly true when the
adverse event could be considered part of the disease being treated (e.g.,
asthma exacerbations occurring with inhalation treatments of asthma).
Ideally, a safety database (and, indeed,
the efficacy database) would include a diverse population in phase 3
studies, and only patients with obvious contraindications would be
excluded from study entry. Inclusion of diverse populations would allow
for the development of safety data in important demographic groups
commonly excluded from clinical trials in the past, such as the elderly
(particularly the very old), patients with concomitant diseases, or
patients taking common concomitant medications. Broadening inclusion
criteria in the studies could enhance the sponsor's ability to generalize
findings to the population likely to use the product in the postmarketing
period.
3. Development of safety (and
effectiveness) data over a range of doses (and plasma levels) throughout
the clinical program
These data help to define the
exposure-response relationship as it relates to safety and effectiveness.
Using a range of doses in phase 3 trials would better characterize the
relationship between exposure and the resulting clinical benefit and risk,
allowing provision of the best dosing advice. (Labeling for doses in
excess of what is needed for effectiveness resulting from inadequate dose
exploration increases risk with no potential for gain.) In addition,
exposure-response data from clinical trials could provide critical
information on the need for dose-adjustments in special populations.
Finally, demonstrating a dose-response relationship in late phase clinical
trials also could add important information to the assessment of efficacy.
Clinical pharmacology studies do not
guarantee a full understanding of all possible risks related to
interactions. Ideally, then, risk assessment would address a number of
potential interactions either during controlled safety and effectiveness
trials or in specific safety trials. This examination for unanticipated
interactions should consider the potential for:
1. Drug-drug interactions, particularly
with likely concomitant medications (e.g., for a new cholesterol
lowering treatment, examining concomitant use of HMG CoA reductase
inhibitors and/or binding resins) and/or products known to interfere
with the metabolism of the investigational product
2. Product-demographic relationships, by
ensuring sufficient diversity of the population (including gender, age,
race, genetics)
3. Product-disease interactions by
ensuring sufficient variability in disease state and likely concomitant
diseases
4. Product-food interactions. This is
particularly important when food effects are seen in PK studies and/or
metabolism data suggest a likelihood of food effects (e.g., CYP3A4
metabolism, a P-glycoprotein pathway, or when food changes
bioavailability)
5. Product-dietary supplement
interactions for commonly used supplements that are likely to be
co-administered or for which reasonable concerns exist
One important way unexpected relationships
can be detected is by incorporating pharmacokinetic assessments (e.g.,
population PK studies) in clinical trials, including in safety trials.
Including PK assessments allows for the determination of exposure-response
relationships for both safety and efficacy (e.g., identifying
unanticipated new interactions or safety issues or confirming the lack
thereof). In addition, such data would allow for better assessment of
whether there is a PK contribution underlying any rare, serious, and
unanticipated adverse events seen in the clinical trials.
When a product has pertinent safety
biomarkers, the markers would be studied during the PK studies and
clinical development (e.g., creatine phosphokinase assessments would be
used in the evaluation of new HMG CoA reductase inhibitors as a marker for
rhabdomyolysis, assessment of QT/QTc effects). If a product has no
acceptable safety biomarker, its clinical trials could be used to develop
and validate such a marker (though such development would not be generally
expected). Although the same dataset would not appropriately be used to
both validate and assess the use of the new marker, development and
validation of biomarkers during clinical trials could be useful in future
trials to address questions regarding product safety.
While comparative safety trials (i.e.,
trials that incorporate an arm with a well-characterized agent, in
addition to the test product) are not generally required in development
programs for novel products,5 such
studies could be useful in the following cases:
1. When there is a need to characterize
background rates of certain adverse events in order to adequately assess
the product
2. When there is a well-established,
well-characterized product with minimal toxicity to treat the condition
of interest. This examination would be intended to show that the novel
therapy has a comparably benign safety profile.
3. When there is a well-established
related therapy. This examination could show whether the toxicity
profile for the established therapy holds for the novel therapy, or
whether important differences exist.
4. When there is a well-established
treatment with an effect on survival or irreversible morbidity.
5. When the sponsor hopes to claim
superiority. In this case, it would normally be expected that such
comparative superiority claims would be based on more than one
controlled study.6
E. What are some special
considerations for optimal risk assessment during product development?
As mentioned above, good risk assessment
practices can vary depending on the product and situation. The following
are examples of how risk assessment strategies could be tailored to suit
special situations.
1. If a product is chronically-used (and
particularly when it has a very long half-life) or has dose-related
toxicities, an examination of whether a maintenance dose lower than the
initial dose or decreases in dosing frequency from the initial
recommended schedule would be appropriate.
2. If a product is to be dose-titrated,
data would be developed to define how titration should be performed and
what the effects of the titration are on safety (and efficacy).
3. If appropriate, an assessment would
be performed of less obvious adverse effects that might not be detected
or readily reported by patients (e.g., effects on cognitive function,
motor skills, sexual function, mood). These assessments could entail the
use of specific psychometric or other validated instruments.
4. If the product is to be studied in
pediatric patients, special safety issues would be considered (e.g.,
growth, neurocognitive development, safety of excipients, universal
immunization recommendations and school entry requirements for
immunization).
5. In certain circumstances, a large,
simple, safety study (LSSS) would be conducted prior to approval. A LSSS
is a clinical study designed to assess relatively few outcomes in a
large number of patients. These outcomes may be important safety
endpoints or other outcomes of clinical importance. Circumstances where
an LSSS would be appropriately considered include:
- When there is a safety signal of
concern in the clinical trial database that is not otherwise well
answered by the available data or likely to be addressed by
remaining outstanding studies (e.g., hepatotoxicity, QT
prolongation).
- When the sponsor is seeking use of
the product as a preventative in asymptomatic individuals. The LSSS
would be intended to assess the background risk, the effectiveness
of the treatment, and the safety of the treatment.
- When there are early signals of
serious toxicities or other unique or special considerations (e.g.,
the safety of the use of the product with a concomitant medication).
In such cases, the LSSS data could either confirm the magnitude and
consequences of any such issues occurring, or show that such
concerns are unfounded.
6. A sponsor could consider reserving
blood samples (or any other bodily fluids/tissues that may be collected
during clinical trials) from some or all patients in phase 3 studies for
possible retrospective testing for various biologic assessments,
including pharmacogenomic markers, immunogenicity, or other biomarkers.
Reserved samples could also allow for retrospective assessment of more
routine tests not prospectively conducted. In particular, having samples
available for retrospective analysis of pharmacogenomic markers could
help to link the occurrence of serious adverse events to particular
genetic markers (e.g., haplotypes). However, if a sponsor were to choose
to retain samples, appropriate informed consent and ethical
considerations would apply.
F. How can sponsors minimize
medication errors?
Ideally, a sponsor would conduct a risk
assessment to ensure that a product's proprietary name, established name,
container label, carton labeling, package insert, and/or packaging do not
inadvertently contribute to medication errors. For example, a sponsor
could perform a medication error prevention analysis or MEPA to:
1. Identify known and potential
medication error modalities
2. Identify potential and actual causes
of each error
3. Prioritize the errors according to
the expected outcomes
4. Minimize the potential for an error
through corrective action including renaming, relabeling or repackaging
Ideally, to assess a product's name,
labeling and packaging, a sponsor would:
1. Obtain first-hand information from
physicians, pharmacists, nurses, and consumers in inpatient and
outpatient settings
2. Use questionnaires, on-duty
observations, interviews, simulation testing, computer models, expert
panels or focus tests
Although FDA currently undertakes such
activities, it would help to minimize medication errors if sponsors also
engaged in such risk assessments to support their proposed names, labeling
and packaging. Further, having such data from the sponsor could help speed
FDA's review of these issues.
We recommend that the potential for the
following serious safety effects be assessed as a part of all new drug
development programs:
1. QTc prolongation
2. Liver toxicity
3. Drug-drug interactions
4. Polymorphic metabolism
For new biological products, we recommend
that the following serious safety effects be assessed:
1. Therapeutic products - immunogenicity,
neutralizing antibodies
2. Biologic products that are live
agents - virulence, trasmissibility, genetic stability
3. Transplantation therapies - survival,
function, host immunocompetence
IV. IMPORTANT CONSIDERATIONS FOR DATA
ANALYSIS AND PRESENTATION
Performing appropriate analyses of safety
data acquired from clinical trials is essential to understanding a
product's risk profile. Many aspects of data analysis and presentation
have been previously addressed in guidance, most notably in FDA's Guideline
for the Format and Content of the Clinical and Statistical Sections of an
Application and the ICH guideline for industry E3 Structure and
Content of Clinical Study Reports. This concept paper does not repeat
these guidances, but presents FDA's thoughts on selected issues for public
discussion and comment.
Although it is important to consider
investigators' descriptions of adverse events, analysis of the whole
safety database requires use of common terminology. In general, sponsors
should utilize one coding convention/dictionary throughout a clinical
program (e.g., Medical Dictionary for Regulatory Activities or MedDRA).
Generally, as an initial approach to data analysis, adverse events can be
examined as they were originally coded. However, specific adverse effects
or toxicities (particularly those with a constellation of symptoms, signs
or laboratory findings) may be reflected by multiple coding terms. When
analyzing an adverse event, sponsors should consider the following:
1. By combining related coding terms, it
is possible both to amplify weak safety signals, and obscure important
toxicities. For example, the constellation of dyspnea, cough, wheezing,
and pleuritis might provide a more sensitive, although less specific,
appraisal of pulmonary toxicity than any single term. Conversely,
combining terms could mask serious, unusual events with more common less
serious events (e.g., constipation might include toxic megacolon).
2. It is important to be aware of the
possibility that coding methods can divide the same event into many
terms. Dividing adverse event terms can decrease the apparent incidence
of an adverse event (e.g., including pedal edema, generalized edema, and
peripheral edema as separate terms could obscure the overall finding of
fluid retention).
Whenever possible, we recommend that the
sponsor, in consultation with FDA, prospectively group adverse event terms
and develop case definitions. A prospective approach is particularly
important for syndromes that are not well characterized by a single term
(e.g., serotonin syndrome, Parkinsonism, drug withdrawal). We recognize,
however, that some groupings can only be constructed after the safety data
are obtained.
Analyzing temporal associations between
product exposure and adverse events is critical to risk assessment,
because it can provide important clues for determining whether the event
was product-related.
Time-to-event analyses are appropriate
for:
1. Clinically important events that
occur on a delayed basis (i.e., events for which even a single
occurrence would be important). For example, progression of disability,
development of cardiac toxicity, and the need for surgical intervention
would be analyzed.
2. Adverse events that occur at
initiation of treatment but diminish in frequency over time (e.g.,
flu-like symptoms with interferons)
Suggested methods for time-to-event
analyses include:
1. Descriptions of risk as a function of
duration of exposure, or as a function of time since initial exposure,
as appropriate (i.e., life table analyses for cumulative incidence)
2. Assessment of risk within discrete
time intervals over the observation period (i.e., a hazard rate curve)
to illustrate the change in risk over time
3. For events found to be associated
with the initiation of treatment that decrease in frequency over time,
we suggest supplemental analyses to attempt to discriminate the relative
contributions of adaptation tolerance, dose reduction, symptomatic
treatment, decreases in reporting, and patient drop out
C. How can analyses of dose effects
contribute to risk assessment?
The relationship between adverse events
and exposure may help determine whether an event is actually related to
the product and, if so, the magnitude of the risk.
Analyses of event rate and severity by
dose should be conducted for clinically important adverse events that may
be drug-related or that would be expected based on pharmacologic class or
pre-clinical data. If there is a range of doses studied, administered dose
is the most common way to assess dose-response, but it may be useful to
look at rate by weight- or body surface area-adjusted dose, especially if
most patients are given the same dose regardless of weight or size. For
products administered over prolonged periods, it may be useful to analyze
event rates based on cumulative dose. When specific demographic subgroups
may be at particular risk of incurring adverse events, exploration of
dose-response relationships by demographic subgroup is important. In
addition:
1. Although the most reliable
information on dose response comes from randomized fixed dose, dose
response studies, potentially useful information may emerge from
titration studies and from attempts to relate adverse events to plasma
concentrations or duration of use.
2. It may also be useful to assess the
relation of adverse event rates to the actual doses received preceding
the events and to assess adverse events by the cumulative dose at the
time of the adverse event.
For products with a stepped dosing
algorithm (i.e., incremental dosing based on age or weight), the actual
cut points of the paradigm are often arbitrary in nature. It may be useful
to make a specific effort to examine safety just above and below the cut
points. For example, if the dose of a product is to be 100 mg for patients
<80 kg and 150 mg for patients _80 kg, an assessment of the comparative
safety profiles of patients 75 to 79.9 kg, versus patients 80 to 84.9 kg
would be valuable.
Data pooling refers to the meta-analysis
of individual patient data (i.e., retrospectively combining patient-level
data from different clinical studies to assess a safety outcome of
interest). Used appropriately, pooled analyses can:
1. Allow detection of relatively rare
events.
2. Enhance the power to detect a
statistical association and protect against chance findings in
individual studies.
3. Provide more reliable estimates of
the magnitude and constancy of risk over time.
However, a negative result from a pooled
analysis does not prove an absence of risk, because the studies may
consist of heterogeneous patient populations, and the methods for
detecting safety outcomes of interest may not be consistent across the
studies. Therefore, data pooling without close attention to the individual
studies may diminish the statistical association and the apparent
magnitude of the risk.
Generally, an appropriately pooled
analysis would have the following characteristics:
1. Phase 1 pharmacokinetic and
pharmacodynamic studies would be excluded.
2. The risk of the safety outcome of
interest would be expressed in person-years, or a time-to-event analysis
would be conducted.
3. The patient population in the pooled
analysis would be relatively homogeneous with respect to such factors as
underlying illness and the studies would have used similar methods of
adverse event ascertainment. Alternatively, subgroup analyses would be
conducted for patients with different baseline or disease
characteristics. Such characteristics could include the disease being
treated and disease severity, gender, age, and/or geographic
location (particularly US vs. non-US sites).
4. A study-specific incidence rate would
be calculated and compared for any signs of case ascertainment
differences (recognizing that study to study variation is to be
expected).
When the results of a pooled analysis show
a diminished statistical association and/or less risk compared to the
safety signal originally obtained from one or more of the contributing
clinical trials, it could suggest inappropriate use of data pooling. If
this occurs, it would be important to ensure that the previously mentioned
principles have been appropriately considered in the analysis.
Demographic subgroup analyses are required
by regulation and other analyses (e.g., effects in people on various
background therapies) are also of interest. Subgroup analyses are, like
most safety analysis to some degree, almost always exploratory, but can
nonetheless be critical in risk assessment. They have the potential
to provide a more reliable and relevant estimate of risk for important
subgroups of the target patient population.
The handling of missing data presents
well-known challenges in data interpretation and presentation. Although
existing guidances discuss this issue, particularly as it applies to
efficacy,7 FDA would be
interested in public comment on ways this issue affects risk assessment
and/or unique methods that could be used to address the challenge that
missing data presents.
FDA and ICH have provided extensive
guidance regarding the presentation of safety data. 8,9
We would supplement these guidances by recommending that certain data be
presented for important adverse reactions with emphasis on the following:
1. Relationship of exposure time to the
development of the adverse event
2. Summary of adverse event rates using
a range of more restrictive to less restrictive definitions (e.g.,
myocardial infarction versus myocardial ischemia)
3. Summary of the distribution of
important demographic variables across the pooled data
4. Where complete case report forms are
called for [21 CFR 314.50], there should also be included hospital
records, autopsy reports, biopsy reports, and radiological reports,
where applicable
5. Assuring that narrative summaries
include important supplementary data (e.g., pertinent lab data, ECG
data, biopsy data), as previously articulated in guidance.9
1 For ease of
reference, this concept paper uses the terms product and drug
to refer to all products (excluding blood products other than plasma
derivatives) regulated by the Center for
Drug Evaluation and Research (CDER) and the Center
for Biologic Evaluation and Research (CBER). Similarly, for ease of
reference, this concept paper uses the term approval to refer to
both drug approval and biologic licensure.
2 FDA and ICH
guidances on clinical pharmacology and preclinical programs are available
at
http://www.fda.gov/cder/guidance/index.htm
and
http://www.fda.gov/cber/guidelines.htm.
3 FDA and ICH
guidances on clinical development are available at
http://www.fda.gov/cder/guidance/index.htm
and
http://www.fda.gov/cber/guidelines.htm.
4 See
Guideline for industry: E1A The Extent of Population Exposure to Assess
Clinical Safety: For Drugs Intended for Long-term Treatment of
Non-Life-Threatening Conditions.
5 Important
exceptions to this general principle exist. For instance, the collection
of comparative safety data is standard practice for some products, such as
new preventive vaccines.
6 See
Guidance for industry: Providing Clinical Evidence of Effectiveness for
Human Drug and Biological Products.
7 See
Guidance for industry: E9 Statistical Principles for Clinical Trials
8 See Guideline
for the Format and Content of the Clinical and Statistical Sections of an
Application.
9 See
Guideline for industry: E3 Structure and Content of Clinical Study
Reports
 Back to Top 
Risk Management Public Workshop
FDA/Center for Drug Evaluation and Research
Last Updated: March 7, 2003
Originator: OTCOM/DLIS
HTML by SJW |
