Determining the Extent of Safety Data Collection Needed in Late Stage Premarket and Postapproval Clinical Investigations

Administrative guidance, not enacted legislation. This captured publisher edition preserves the listed document and attachments; later developments are not independently consolidated.

Determining the Extent
of Safety Data
Collection Needed in
Late-Stage Premarket
and Postapproval
Clinical Investigations
Guidance for Industry
U.S. Department of Health and Human Services
Food and Drug Administration
Center for Drug Evaluation and Research (CDER)
Center for Biologics Evaluation and Research (CBER)
February 2016
Clinical/Medical

Determining the Extent of
Safety Data Collection
Needed in Late-Stage
Premarket and Postapproval
Clinical Investigations
Guidance for Industry
Additional copies are available from:
Office of Communications, Division of Drug Information
Center for Drug Evaluation and Research
Food and Drug Administration
10001 New Hampshire Ave., Hillandale Bldg., 4th Floor
Silver Spring, MD 20993-0002
Phone: 855-543-3784 or 301-796-3400; Fax: 301-431-6353
Email: druginfo@fda.hhs.gov
http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/default.htm
and/or
Office of Communication, Outreach and Development
Center for Biologics Evaluation and Research
Food and Drug Administration
10903 New Hampshire Ave., Bldg. 71, Room 3128
Silver Spring, MD 20993-0002
Phone: 800-835-4709 or 240-402-8010
Email: ocod@fda.hhs.gov
http://www.fda.gov/BiologicsBloodVaccines/GuidanceComplianceRegulatoryInformation/Guidances/default.htm
U.S. Department of Health and Human Services
Food and Drug Administration
Center for Drug Evaluation and Research (CDER)
Center for Biologics Evaluation and Research (CBER)
February 2016
Clinical/Medical

Contains Nonbinding Recommendations
TABLE OF CONTENTS
I. 	INTRODUCTION............................................................................................................. 1
II. 	BACKGROUND ............................................................................................................... 2
III. 	SELECTIVE SAFETY DATA COLLECTION ............................................................ 3
A. 	Types of Clinical Investigations That May Be Considered for Selective Safety Data
Collection......................................................................................................................................... 4
B. 	Types of Safety Data That May Be Appropriate for Selective Safety Data Collection ........... 5
C. 	Other Considerations for Safety Data Collection ....................................................................... 6
D. 	Types of Safety Data That Should Always Be Collected ............................................................ 6
IV. 	METHODS FOR SELECTIVE SAFETY DATA COLLECTION ............................. 7
V. 	AGREEMENT BETWEEN SPONSORS AND FDA ON A PLAN FOR SELECTIVE
SAFETY DATA COLLECTION ......................................................................................... 8
i

Contains Nonbinding Recommendations
Determining the Extent of Safety Data Collection Needed in Late-
Stage Premarket and Postapproval Clinical Investigations
Guidance for Industry1
This guidance represents the current thinking of the Food and Drug Administration (FDA or Agency) on
this topic. It does not establish any rights for any person and is not binding on FDA or the public. You
can use an alternative approach if it satisfies the requirements of the applicable statutes and regulations.
To discuss an alternative approach, contact the FDA office responsible for this guidance as listed on the
title page.
I. 	INTRODUCTION
This guidance is intended to help sponsors determine the amount and types of safety data to
collect during late-stage premarket and postapproval clinical investigations, (e.g., phase 3
clinical trials, studies of new uses, long-term outcomes). This guidance discusses a selective
approach to safety data collection during late-stage premarket development or during the
postapproval stage based on what is already known about a drug’s2 safety profile. This guidance
provides recommendations on when to consider selective safety data collection and how to do so
to maintain a balance between eliminating the collection of data that will not be useful and
collecting sufficient data to allow adequate characterization of the safety profile of a drug. In
addition, this guidance provides information to sponsors about consulting with the relevant FDA
review division or divisions to determine whether a selective approach to safety data collection
would be appropriate (see section V of this guidance).
This guidance is intended to apply to safety data collection during late-stage premarket and
postapproval clinical investigations in all disease settings except rare diseases (see section III of
this guidance). There is an existing FDA guidance that applies to the collection of clinical data
in all phases of oncology clinical trials,3 but that guidance does not discuss selective safety data
1 This guidance has been prepared by the Office of Medical Policy in the Center for Drug Evaluation and Research
in cooperation with the Center for Biologics Evaluation and Research at the Food and Drug Administration.
2 For the purposes of this guidance, all references to drugs or drug products include both human drugs and
biological drug products regulated by the Center for Drug Evaluation and Research or the Center for Biologics
Evaluation and Research unless otherwise specified.
3 See the guidance for industry Cancer Drug and Biological Products – Clinical Data in Marketing Applications.
We update guidances periodically. To make sure you have the most recent version of a guidance, check the FDA
Drugs guidance Web page at
http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/default.htm.
1

Contains Nonbinding Recommendations
collection in the specific instances covered in this guidance. For this reason, the
recommendations in this guidance may differ from those in the oncology-focused guidance.
Where differences occur, the recommendations in this guidance apply.
FDA is also aware that some of the recommendations in this guidance may not align with the
expectations of safety data collection in other regions or countries, which may lead to difficulty
in implementing this guidance in some clinical investigations. However, we believe this
guidance will give sponsors the flexibility to design and implement protocols with selective
safety data collection where appropriate.
This guidance is not intended to affect reporting (as opposed to collection) of postmarketing
adverse events relevant to an approved drug as required under 21 CFR 314.80 and 600.80 or
affect reporting of investigational new drug application (IND) safety information as required
under § 312.32 (21 CFR 312.32).4 Those reporting requirements remain unchanged, and
selective safety data collection may only occur in a manner that would permit all regulatory
reporting requirements to be fulfilled.
In general, FDA’s guidance documents do not establish legally enforceable responsibilities.
Instead, guidances describe the Agency’s current thinking on a topic and should be viewed only
as recommendations, unless specific regulatory or statutory requirements are cited. The use of
the word should in Agency guidances means that something is suggested or recommended, but
not required.
II. 	BACKGROUND
A robust safety database is critically important to accurately assess and adequately characterize
the risks of a new drug. Sponsors collect extensive safety-related data throughout the course of
drug development, and knowledge about a drug’s safety profile continually evolves as safety
data accumulate. Comprehensive safety data, including essentially all adverse events, are
collected in the early stages of drug development. In the later stages of premarket development
or in postapproval studies or clinical trials, it may be appropriate to use a selective approach to
safety data collection for common, non-serious adverse events that have already been well-
characterized through data collection in earlier stages. For example, if safety data already
collected on hundreds of patients indicate that 17 percent reported a headache when on drug
treatment compared with 10 percent on placebo, collection of similar data in thousands of
additional patients in a large phase 3 trial would minimally refine this value and would require
extensive resource utilization, while providing no important new information. In this situation, a
limited collection of safety data may be appropriate.
4 FDA’s final rule discusses the rationale for more-selective use of IND safety reports (Investigational New Drug
Safety Reporting for Human Drug and Biological Products and Safety Reporting Requirements for Bioavailability
and Bioequivalence Studies in Humans (75 FR 59935, September 29, 2010)).
The guidance for industry and investigators Safety Reporting Requirements for INDs and BA and BE Studies also
discusses FDA’s thinking on reporting information that is interpretable and that will meaningfully contribute to a
drug’s safety profile. (IND means investigational new drug application; BA and BE means
bioavailability/bioequivalence).
2

Contains Nonbinding Recommendations
Encouraging selective safety data collection in late-stage premarket and postapproval clinical
investigations is consistent with FDA’s overall approach to safety assessment, which focuses on
information that is useful and adds to existing knowledge. For example, § 312.32 requires
expedited reporting of serious, unexpected suspected adverse reactions to FDA and all
investigators during drug development.
In some cases, collection of data that do not contribute to better characterizing the safety profile
of a drug may have negative consequences. For example, there is growing interest in larger,
simpler trials to obtain outcome data, data on long-term effects of drugs, and data on
comparative effectiveness and safety. Excessive safety data collection may (1) discourage the
conduct of these types of trials by increasing the resources needed to perform them and (2) be a
disincentive to investigator and patient participation in clinical trials. Thus, selective safety data
collection may (1) facilitate the conduct of larger trials without compromising the integrity and
the validity of trial results or losing important information, (2) facilitate investigators’ and
patients’ participation in clinical trials, and (3) help contain costs by making more-efficient use
of clinical trial resources. For these reasons, selective safety data collection should be
considered in circumstances where it is appropriate (see section III of this guidance).
III. 	SELECTIVE SAFETY DATA COLLECTION
If the drug’s safety profile has already been well-established, selective safety data collection may
be appropriate during late-stage premarket and postapproval clinical investigations. Examples of
selective safety data collection include, but are not limited to, the following:
 	No collection of certain safety data
 	Less-frequent collection of certain safety data
 	Collection of certain safety data from only a fraction of the total trial enrollment (e.g.,
10 percent of patients in a large trial)
In general, selective safety data collection may be appropriate for certain types of safety data
when the following conditions are met:
 	The number of patients and their characteristics, the duration of exposure, and the
dose range used in previous clinical investigations are sufficient to adequately
characterize the safety profile of the drug for common, non-serious adverse events.
 	The occurrence of common, non-serious adverse events has been generally similar
across multiple clinical investigations.
 	The drug’s safety profile is established to the extent that it is reasonable to conclude
that the occurrence of common, non-serious adverse events in the population to be
studied will be similar to rates observed in previously conducted clinical
investigations.
3

Contains Nonbinding Recommendations
For clinical investigations in which these conditions are met, the amount and types of data
appropriate for selective safety data collection will vary depending on a range of factors,
including the disease being studied, the patient population, subgroup or subgroups of interest,
nonclinical findings, prior experience with the drug and drug class, and study design.
A. 	Types of Clinical Investigations That May Be Considered for Selective Safety
Data Collection
Selective safety data collection may be appropriate in the following types of clinical
investigations. However, these are not the only circumstances where selective safety data
collection may be appropriate, and this is not intended to be a complete list.
 	Clinical investigations of new indications of approved drugs: Selective safety
data collection may be appropriate in clinical investigations for new indications if the
existing safety database is relevant to such investigations, including, for example, the
similarity of the population studied in the new investigation to patients previously
studied and the doses used.
 	Postapproval clinical studies and trials conducted to fulfill postmarketing
requirements and postmarketing commitments: When the study population in
these types of studies or trials is generally the same as or is similar to the population
from which the premarket safety database was derived, safety data collection can
often be limited to the primary safety endpoint and other endpoints of interest.
 	Late-stage premarket and postapproval outcome clinical trials: Outcome clinical
trials usually involve relatively large populations (often 5,00015,000 patients)
similar to populations previously studied and are usually of longer duration than
earlier studies. Typically, a fairly substantial safety database is in existence before
initiation of an outcome clinical trial, and this database already provides sufficiently
precise estimates of common, non-serious adverse events (e.g., those that occur in 5
percent or more of exposures). Even when there is not extensive prior safety data,
full data could be collected in a subset of patients (e.g., 1,000) with more-selective
data collection in other patients.
 	Premarket clinical investigations for some original applications: Unless
sufficient safety data already exist to adequately characterize the safety profile of a
drug, comprehensive safety data collection is expected throughout premarket clinical
development intended to support approval of a new drug application (NDA) or a
biologics license application (BLA) for a novel agent in order to elucidate occurrence,
dose-response, and subset (demographic, concomitant illnesses, concomitant therapy)
variations for the full range of adverse events of the drug. However, even in a
development program for a novel agent, if sufficiently comprehensive safety data
become available before completion of clinical development, selective safety data
collection in some late-stage investigations may be appropriate. For example,
selective safety data collection may be appropriate in phase 3 trials where the existing
4

Contains Nonbinding Recommendations
safety database adequately characterizes a drug’s safety for more common, non-
serious adverse events; therefore, data on those events may not need to be collected.
 	Postapproval clinical investigations in a different patient population or with
different doses or other conditions of use: Selective safety data collection
generally may not be appropriate in clinical investigations of marketed drugs in which
there are important differences in the patient population, dose, dosage regimen,
duration of use, or route of administration compared with the conditions of use for the
marketed indications. For example, the safety database of a drug approved to treat
active cancer generally would not be adequate to permit selective safety data
collection in an investigation of the long-term use of that drug for adjuvant therapy or
in the prevention of cancer in a healthy population. Even in this setting, however, it
may be appropriate, in some cases, to collect common, non-serious adverse events in
only a subset of the overall study population. An exception may also apply in cases
where a lower dose or shorter duration of therapy is being investigated.
B. 	Types of Safety Data That May Be Appropriate for Selective Safety Data
Collection
Where selective safety data collection is suitable, it may be appropriate to limit collection of data
or stop collection of certain types of data, including the following:
 	Non-serious adverse events not associated with dose modification, drug
discontinuation, or withdrawal from the trial: Generally, these events are well-
characterized when a drug is far along in development or has obtained marketing
approval, and additional data collection would be unlikely to add to the established
safety profile of a drug.
 	Routine laboratory monitoring: In many cases, it may be possible to eliminate
routine laboratory monitoring or to decrease the frequency of monitoring certain
laboratory parameters (e.g., less-frequent liver function testing as safety profile data
diminish concerns about hepatotoxicity).
 	Information on concomitant medications: If existing data satisfactorily
characterize all anticipated drug-drug interactions and metabolic pathways, additional
detailed information on concomitant medications (e.g., dose, dosing schedule, start
and stop dates) may be of limited use, particularly for drugs used only short term.
 	Patient history and physical exams: Less-detailed histories and less-frequent
physical exams for patients may be appropriate in some circumstances, especially in
outcome clinical trials. If an abbreviated history is considered appropriate but certain
aspects of the history are considered relevant and should be collected (e.g.,
cardiovascular disease, psychiatric history, including history of depression,
suicidality, or both), these aspects should be described in the protocol.
5

Contains Nonbinding Recommendations
C. 	Other Considerations for Safety Data Collection
For situations that may be appropriate for selective safety data collection, sponsors should
nonetheless consider where there may be reasons to collect complete data to better characterize
the safety profile of a drug. This may apply even for adverse events that are not serious, fatal, or
dose-modifying or that do not cause drug discontinuation or withdrawal from the trial. For
example:
 	Collection of complete safety data in population subsets: If availability of data in
a subset of a population is limited, generally all data on use of a drug in these specific
populations should be collected (including data on non-serious adverse events). For
example, information on exposure in pediatric patients, pregnant and nursing women,
and geriatric populations is often limited, so that selective safety data collection may
not be appropriate in those populations.
 	Collection of complete safety data to identify risk factors: The incidence and
severity of an adverse event may be related to baseline factors and other aspects of
treatment. For this reason, where there is a known or suspected causal relationship
between a drug and an adverse event, it may be important to continue collecting
complete safety information throughout drug development to better characterize the
relationship and the effect of these influences (e.g., baseline risk or demographic
factors, dose-response, drug concentration, concomitant therapy, concomitant illness)
on risk.
D. 	Types of Safety Data That Should Always Be Collected
Sponsors should be aware that the following types of data are generally not appropriate for
selective safety data collection and should always be collected:
 	Data on all serious adverse events.
 	Data on non-serious adverse events that lead to dose modification, drug
discontinuation, or withdrawal from the trial.
 	Data on unscheduled study visits, hospitalizations, and accidental injuries because
these events may reflect serious adverse events of the drug.
 	In an oncology setting, data from all Grade 3 and Grade 4 adverse events, as well as
Grade 2 adverse events that affect vital organs (e.g., heart, liver). 5,6
5 The grading system was used for adverse events in cancer trials described in the Common Terminology Criteria for
Adverse Events document from the National Cancer Institute (U.S. Department of Health and Human Services’
National Cancer Institute at the National Institutes of Health). The grading system is accessible at
http://evs.nci.nih.gov/ftp1/CTCAE/About.html (under the heading “Files: Booklet”).
A Grade 4 adverse event results in the need for life-saving and urgent intervention.
6

Contains Nonbinding Recommendations
 	In development programs for rare disease indications, complete safety data should be
collected, because these trials generally have limited patient populations and are
unlikely to meet the recommendations for selective safety data collection.
For these types of safety data, it is generally important to collect information on all occurrences
to better understand the following:
 	Causality
 	Incidence
 	Severity of adverse events
 	Populations that are at risk
 	Dose-response
 	Other factors that contribute to our understanding of the nature of the event and who
is at risk
IV. 	METHODS FOR SELECTIVE SAFETY DATA COLLECTION
Apart from not collecting certain safety data, a plan for selective safety data collection could
include the following modifications as appropriate:
 	Limiting safety data collection to a pre-identified subset of study population: It may
be appropriate to collect certain safety data (e.g., non-serious adverse events) from a
sample of the study population. Methods for selecting the subpopulation should be
detailed in the protocol. These may include selecting patients randomly, selecting study
sites randomly, or selecting only the larger population sites. When selecting patient
subsets or sites, it may be important to ensure representation from important demographic
or disease-specific subgroups.
 	Decreasing the frequency of data collection: Less-frequent collection of specific safety
data, such as laboratory data and physical examination data, may be appropriate for long-
term trials of drugs with substantial short-term exposure. For example, laboratory studies
and physical-examination data may be collected at relatively long intervals (e.g., every 6
months). Telephone and online follow-up between study visits may help maintain
contact with patients.
Specific safety data that will not be collected for the selective collection plan should be listed in
the protocol.
A Grade 3 adverse event results in severe or clinically significant, but not immediately life-threatening,
consequences; results in hospitalization or prolonged hospitalization; disables an individual; and results in
limitations on ability to bathe, dress, undress, feed, use the toilet, and take medications.
A Grade 2 event results in the need for moderate, minimal, or local or invasive intervention and results in limitations
on age-appropriate ability to prepare meals, shop for groceries or clothes, use the telephone, and manage money.
6 The guidance for industry Gene Therapy Clinical Trials – Observing Subjects for Delayed Adverse Events
discusses collecting delayed adverse event data in long-term follow-up for gene therapy trials.
7

Contains Nonbinding Recommendations
V. 	AGREEMENT BETWEEN SPONSORS AND FDA ON A PLAN FOR
SELECTIVE SAFETY DATA COLLECTION
A sponsor considering selective safety data collection should consult with the relevant FDA
review division or divisions to determine whether such collection would be considered
appropriate, and if so, develop its plan for implementation. The sponsor should discuss its
specified plan with the relevant FDA review division or divisions at the appropriate time (e.g., at
the end-of-phase 2 meeting for selective safety data collection for a phase 3 trial) to determine
whether the plan is acceptable and to reach an agreement with the division or divisions on the
details of the plan. The agreement should be incorporated into the procedures for safety data
collection in the protocol, the monitoring plan, and other appropriate trial documents. These
steps should help ensure consistent implementation across trial sites, help support risk-based
monitoring efforts, and help alleviate potential inspectional problems related to safety data
collection.
8