Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 3
Summary
Prescription drugs1 provide great benefit to the public’s health. However,
most drugs also pose risks to health, and often these risks cannot be identified or
fully characterized until after a drug has entered the marketplace. Because the US
Food and Drug Administration (FDA) is the agency responsible for ensuring that
the benefits of a prescription drug outweigh its risks, the timely identification of
and response to risks from marketed drugs are central to its mission. Before 2007,
FDA’s options in responding to risks of concern that emerged postmarketing were
limited. It could either withdraw a drug from the market altogether or negotiate
with sponsors to get them to accept a change in its regulatory status with regard
to labels, warnings, and the like. The FDA Amendments Act of 2007 (Public Law
[PL] 110-85; FDAAA) provided FDA with new postmarketing regulatory tools to
better protect the health of the public, including the authority to require an indus -
try sponsor to conduct a clinical trial or other research study in the postmarketing
setting (called postmarketing requirements).2 That expanded authority brought a
new set of ethical and scientific questions for FDA to consider; FDA asked the
Institute of Medicine (IOM) to conduct a study to address these questions.
1 For simplicity, the committee uses the term drugs throughout this report, but similar considerations
would apply to biologics. The committee’s charge is related to the Food and Drug Administration’s
regulation of the drug and biologics supply. When discussing FDA’s regulatory authority and mission,
therefore, the committee does not address FDA’s roles related to other products, such as tobacco,
medical devices, veterinary medicines, the food supply, and animal feed.
2 A postmarketing requirement (PMR) is an FDA-required research study that a sponsor must
conduct after a drug has been approved and is released on the market.
3
OCR for page 4
4 STUDYING THE SAFETY OF APPROVED DRUGS
CHARGE TO THE COMMITTEE
In April 2010, FDA asked IOM to “convene a committee to evaluate the sci-
entific and ethical issues involved in conducting studies of the safety of approved
drugs”. The five specific questions posed by FDA appear in Box S-1. In response
to FDA’s request, IOM convened a committee of 12 members who have expertise
in bioethics, biostatistics, clinical trials, epidemiology, health policy, law, patient
safety, pharmacoepidemiology, and regulatory science.
FDA requested two reports: a letter report due in July 20103 and this final
report. In its letter report, Ethical Issues in Studying the Safety of Approved
Drugs: A Letter Report, released on July 9, 2010, the committee addressed the
first question of the committee’s charge by presenting a conceptual framework
for analyzing the ethics of postmarketing randomized controlled trials required
by FDA. In this final report, the committee addresses all five specific questions
posed to the committee by FDA.
COMMITTEE’S APPROACH TO ITS CHARGE
The committee met in person six times, including two open information-
gathering sessions. The committee used the conceptual framework in its letter
report as a starting point for this final report but conducted further research and
deliberations related to its full charge. The following underpinnings of that con -
ceptual framework, as well as additional themes that emerged as the committee
deliberated its full charge, shaped this report: (1) an understanding of FDA’s
public health mission; (2) the importance of adopting a lifecycle approach to
drug safety and benefit–risk assessment; (3) FDA’s ethical obligations in making
regulatory decisions, including the centrality of transparency and communication
to those decisions; and (4) a commitment to using best practices in regulatory
science and high-quality evidence in regulatory decision-making.
BENEFIT–RISK ASSESSMENT AND MANAGEMENT
THROUGHOUT A DRUG’S LIFECYCLE
In Chapter 2, the committee explains the need for FDA to adopt a consistent
process for factoring in different evidence in the making of regulatory deci -
sions throughout a drug’s lifecycle. Using a consistent framework for regulatory
decision-making will facilitate stakeholder understanding of decisions and the
process by which decisions are made, emphasize the dynamic nature of benefit
and risk assessments, provide an opportunity to consider the value of additional
3 FDA requested that IOM complete the letter report before a July 13–14, 2010, joint meeting
of FDA’s Endocrinologic and Metabolic Drugs Advisory Committee and Drug Safety and Risk
Management Advisory Committee on rosiglitazone.
OCR for page 5
5
SUMMARY
BOX S-1
Charge to the Committee
The US Food and Drug Administration (FDA) has requested that the
Institute of Medicine convene a committee to evaluate the scientific and
ethical issues involved in conducting studies of the safety of approved drugs.
Questions to be explored by a committee include
1. What are the ethical and informed consent issues that must be con-
sidered when designing randomized clinical trials to evaluate potential
safety risks?
2. What are the strengths and weaknesses of various approaches,
including observational studies, including patient registries, meta-
analyses, including patient-level data meta-analyses, and randomized
controlled trials, to generate evidence about safety questions?
3. Considering the speed, cost, and value of studies, what types of
follow-up studies are appropriate to investigate different kinds of sig-
nals (detected pre-approval or postapproval) and in what temporal
order?
4. Under what circumstances should head-to-head randomized clinical
trials for safety be required?
5. How should FDA factor in different kinds of safety evidence in consid-
ering different kinds of regulatory actions?
postmarketing studies, and minimize delays and increase transparency in the
agency’s decisions. The committee proposes a three-stage framework that FDA
could apply to regulatory decisions prompted by new information that could
affect a drug’s benefit–risk profile, and to use for periodic re-evaluations of the
benefit–risk profile during a drug’s lifecycle.4 Central to the framework is the
importance of eliciting and incorporating the perspective of the patient. 5 In the
first stage (Stage I) of the framework, FDA should define the public health ques -
tion that prompted the need for a regulatory decision, including identifying the
specific characteristics of the drug and health problem at issue, available informa-
tion about the drug, alternative treatments that are available, and plausible regula-
tory actions and their potential consequences. In the second stage (Stage II) of
the framework, FDA should evaluate the quality of evidence on both the benefits
4 Given its charge, the committee focuses on the use of the framework in the postmarketing setting,
but it could also be useful in the premarketing setting.
5 FDA should ensure that patient advocacy groups represent the views of patients rather than the
views of commercial entities that provide funding to the organizations.
OCR for page 6
6 STUDYING THE SAFETY OF APPROVED DRUGS
and the risks associated with the drug, including any new information that has
triggered the need to consider regulatory action. The output of this stage includes
estimates of the likelihood and magnitude of a drug’s benefits and risks and a
characterization of the scientific evidence on which the estimates are based. The
third stage (Stage III) of the framework is the stage in which regulatory decisions
are made and implemented. This stage involves synthesizing and integrating the
estimates of benefits and risks and the quality of the evidence on which these are
based (from Stage II) with the public health question (as specified in Stage I);
deciding on the appropriate regulatory actions, including whether further study
should be required; communicating the decision; implementing the regulatory
actions; evaluating the effects of the regulatory actions; and, particularly in the
case of complex or difficult decisions, evaluating the decision-making process
and the impact of the action taken on the public’s health.
In order to formalize and make concrete a commitment to a lifecycle
approach to drug oversight and benefit–risk management, the committee proposes
that FDA develop and make public a living document, which the committee calls
a Benefit and Risk Assessment and Management Plan (BRAMP), that FDA would
update whenever it re-evaluates a drug’s benefit–risk profile. The document
would serve as a guide that supports organizational adherence to the lifecycle
approach, increases the transparency of FDA’s decisions, and fosters collabora -
tion between FDA and drug sponsors.
In light of its charge, the committee focuses on the ethical and scientific
issues associated with one of the regulatory actions available to FDA: impos -
ing a postmarketing requirement on a drug’s sponsor to conduct additional
research, including studies that employ observational and RCT designs. 6 When
FDA imposes a postmarketing requirement, it is expressing not only scientific
uncertainty about the harms or benefits of a drug but a judgment that the public
health interests served by requiring additional research outweigh the burdens
placed on pharmaceutical manufacturers and—more important from an ethical
standpoint—any risk of harm to or burdens on research participants. Because
it has required the studies, FDA bears a measure of ethical responsibility for
any adverse outcomes that research participants experience. When requiring
postmarketing research, therefore, FDA should be confident that the information
from the study is necessary to answer the public health question that prompted
the consideration of a regulatory action, that the required study is designed and
conducted in such a way that it can provide the necessary information, that it will
use the study findings in a timely manner in its regulatory decisions, and that the
6 In
this report, the committee uses the term studies, in accordance with ordinary usage in the
scientific literature, as a parent or generic term that encompasses research projects of all types
regardless of design. Thus, as the committee uses the term studies, it applies to both clinical trials
and non–clinical-trial investigations, such as observational investigations. When referring specifically
to either observational designs or randomized controlled trial (RCT) designs, the committee uses
those specific terms.
OCR for page 7
7
SUMMARY
design and conduct of the study adequately protect the rights and interests of
research participants.
FDAAA provides FDA the authority to require postmarketing research
under specific circumstances, and FDA has interpreted its authority in guidance
to industry about postmarketing requirements. The committee identifies, more
specifically, circumstances that typically increase the uncertainty surrounding the
benefit–risk profile of a drug under which FDA should seriously consider requir-
ing postmarketing studies and, if it decides not to require research, it should make
public its reasoning for that decision.
EVIDENCE AND DECISION-MAKING
FDA’s drug-related regulatory decisions are based, to a large extent, on
scientific evidence. Statistical inferences must be made in order to use scientific
evidence in decision-making. The two main approaches to statistical inference are
the standard “frequentist” approach and the Bayesian approach. The frequentist
approach to statistical inference is familiar to medical researchers and is the basis
of most FDA rules and guidance. The Bayesian approach is less widely used
and understood, and it has many attractive properties that can both elucidate the
reasons for disagreements and provide an analytic model for decision-making
that can allow decision-makers to combine the chance of being wrong about ben-
efits and risks with the seriousness of those errors to support optimal decisions.
Bayesian, frequentist, and other relevant methods can be useful for integrating
new information about a drug into our current understanding about its association
with harms or benefits to inform decisions, and FDA should develop a sufficient
body of internal expertise of these approaches.
Scientists and technical experts often disagree about the available evidence.
Those disagreements can result from different prior beliefs about the existence of
a given benefit or risk, different opinions about the quality of a new study, differ-
ent opinions about the relevance of the new evidence to the public health ques -
tion, and different ideas about how to synthesize and weigh all evidence relevant
to the public health question. In addition, even if scientists do not make the final
decision about what regulatory action FDA should take, they often have opinions
about what level of certainty there should be for a given regulatory decision, and
that opinion can shape their assessment of the strength of the evidence. Decisions
where scientists and technical experts disagree about the available evidence, such
as in the case of rosiglitazone (Avandia®), are often the most difficult ones that
FDA must make.
If the underlying reasons for disagreements are not properly expressed or
elicited, however, it will be difficult to reach a consensus on the appropriate regu-
latory action. A lack of clarity about those disagreements makes it extraordinarily
difficult for those involved in the decision-making process to understand the
reasons for the disagreements, adjudicate them, and make decisions. The nature
OCR for page 8
8 STUDYING THE SAFETY OF APPROVED DRUGS
of scientific differences should be identified and explicitly stated so that all par-
ticipants in the decision-making process—including scientists, technical experts,
patients and other stakeholders, and the decision-makers themselves—understand
the underlying sources of scientific disagreement. To maintain the public’s trust
in the agency, it is important that the public can understand how FDA weighs the
factors that lead to disagreements in its decisions. FDA’s handling of its recent
regulatory decision to place restrictions on the distribution of rosiglitazone pro -
vides a good example of how it can make public both the disagreements among
agency scientists about evidence and how those disagreements were considered
in the agency’s decision. The three-stage framework discussed above provides a
process for eliciting and resolving those differences, and the BRAMP document
constitutes a mechanism to provide transparency to the public about the disagree-
ments and how they affected the decision.
The reanalyses of data from rosiglitazone trials revealed discrepancies and
judgment calls that occurred from ascertaining clinical events through data
analysis that affected the interpretation of the evidence and, later, the regulatory
decision. That example highlights the importance of adherence to principles of
reproducible research, that is, presenting analyses in such a way that the reader
of results can understand most of or all the process that occurred from the gather-
ing of the data to the reporting of specific analyses. At a minimum, that requires
provision of study protocols with statistical-analysis plans, statistical code, and
information about how decisions were made to produce the analytic dataset from
the raw measured data. Optimally, it involves some form of data sharing. Adher-
ing to those principles, for both premarketing and postmarketing data, would
facilitate a better understanding of FDA’s decisions by the public and allow
analysts both in and outside FDA to investigate the sensitivity of conclusions to
different analytic approaches.
SELECTION AND OVERSIGHT OF REQUIRED
POSTMARKETING STUDIES
As a public health agency, FDA has ethical obligations both to protect the
public from unsafe drugs and to safeguard the rights and interests of participants
in the research that supports the agency’s decisions about drug benefits and risks.
Once it has decided to require postmarketing research, FDA must balance those
two obligations when determining what type of study or studies to require. FDA
may be justified in requiring studies that might expose participants to more net
risk than they would probably face in regular clinical practice, that offer partici -
pants no reasonable expectation of clinical benefit, or both.7 This is only justified
7 Although there may not be any potential clinical benefit from the drug, research participants could
benefit from participation through improved clinical care and being provided at no cost a drug that
they could not afford.
OCR for page 9
9
SUMMARY
when a question of pressing public health importance is at stake, no other design
with a better benefit–risk balance for participants could supply the evidence
needed for a responsible regulatory response to that question, FDA uses the find -
ings of the research in formulating its regulatory response, and special safeguards
are in place to protect the rights and interests of the research participants.
As discussed in Chapter 4, FDAAA allows FDA to require an RCT only
when observational studies are not sufficient to provide the information the
agency needs. The challenge for FDA is to determine when one or more postmar-
keting observational studies are necessary and adequate to inform its regulatory
decision, when one or more postmarketing RCTs are necessary and adequate, or
when some combination of studies are required.
To make that decision, FDA should consider the advantages and disadvan-
tages of observational studies and RCTs in the postmarketing setting, which poses
distinct scientific and ethical challenges. In contrast to the premarket context,
the drug of interest, once approved, will be in general use, making observational
studies feasible, and patients will have access to it without having to participate
in FDA-required research. After approval, evidence about the drug’s benefits and
risks can be generated by multiple investigators, supported by various funders,
and generated from a wide variety of observational studies and RCTs. Concerns
about a drug’s benefit–risk profile are likely to involve comparisons with other
active treatments that are considered to be therapeutic alternatives.
In theory, an ideal RCT could provide the evidence that regulators need to
identify the best regulatory response to a public health question of interest. In
practice, however, a number of constraints can make the ideal trial infeasible
and increase the potential for imprecision, bias, and decreased transportability 8
of results: patients may refuse to participate, making the study population less
representative of the target population taking the drug; adherence may not reach
desired levels; and patients may withdraw from the trial. Other practical consid -
erations can limit an RCT or observational study’s ability to address the public
health question. Those limitations include the ability to measure the endpoints
of primary interest, the determinants of drug choice, timeliness, the quality of
available data sources or easily obtainable information for observational stud -
ies, the frequency of the endpoint being studied, the availability of participants,
and the effect size of interest. The public health question also greatly affects the
choice of design, with RCTs having advantages over observational studies when
the public health question focuses primarily on a drug’s benefits or a relative
effect size is likely to be small; observational studies have some advantages over
8 The term transportability is used in this report, rather than external validity or generalizability,
because the committee thinks that it reflects a nonbinary characteristic better. Different effects can
occur in a variety of settings, and study results may be transportable to some populations or settings
but not others, so transportability may not be a simple binary property.
OCR for page 10
10 STUDYING THE SAFETY OF APPROVED DRUGS
RCTs when the primary focus is on rare or delayed risks, and the relative risks
are moderate or large.
Virtually every study type has tradeoffs dictated by either its design or its
conduct. Regardless of the type of design chosen, the study should have pre-
specified protocols, endpoints, analytic plans, and procedures in place to ensure
adequate and uniform followup, ascertainment and adjudication of endpoints, and
other steps to ensure data quality.
While often methodologically superior to observational studies that use
existing data, in the required postmarketing context a prospective cohort study
that is designed so that followup starts at the initiation of drug treatment can raise
some of the same ethical issues that are raised by an RCT, assuming both occur at
the same time in the drug’s lifecycle. In these circumstances, there may be ethi -
cal advantages to FDA’s requiring observational designs examining the effects
of past exposures, which may yield results sooner and enable faster regulatory
action to protect the public’s health. The ability of such designs to provide the
information FDA requires for regulatory action rests not only on the availability
of high-quality data but also on access to those data. Access is itself a function
of ethical considerations related to privacy and authorization.
In deciding what type of RCT to require, FDA should consider which RCT
would best approximate the ideal hypothetical trial. As with observational stud -
ies, the choice of a comparator for a drug is an important element in the design
of an RCT. If an effective treatment is available for the same indication, an
active-controlled design (a head-to-head trial, defined as a comparison of two
active treatments indicated for the same patients with the same conditions) is
often preferred on both ethical and public health grounds. It may be ethically
acceptable, however, for FDA to require a placebo-controlled postmarketing
trial under some specific circumstances even if an alternative treatment is avail -
able—such as studies of interventions intended to provide symptomatic relief for
minor, self-limiting, or reversible conditions and in short-term trials to evaluate
surrogate endpoints.
When FDA requires postmarketing research, it has an obligation to ensure
that the research is conducted ethically. One component of that obligation is to
ensure that, when appropriate, the study secures the voluntary informed consent
of research participants. However, the ethical obligation to obtain prior informed
consent is not applicable to all required postmarketing research.
There are not always ethically relevant distinctions between some kinds of
observational research that FDA could require manufacturers to conduct and FDA
surveillance activities that are classified as public health practice. It is unclear
whether FDA’s human subjects regulation (21 CFR 50) or the Common Rule (45
CFR 46) is the operative regulation for FDA-required, postmarketing observa-
tional studies, nor is it clear whether some or all types of observational designs
qualify as clinical investigations under 21 CFR 50. It is important that FDA clarify
whether its human subjects regulations govern required postmarketing observa-
OCR for page 11
11
SUMMARY
tional studies and, if so, how FDA will address and expect institutional review
boards (IRBs) to address any differences between 21 CFR 50 and 45 CFR 46 in
oversight and research participant protections for different observational designs.
The desirability of linking datasets and of obtaining additional information
from patients, or otherwise needing access to some identifying information about
patients, will probably increase. That increase could lead to additional ethical
questions about the adequacy of data-security practices, authorization for access
to different datasets, and the difference, if any, between research and public health
goals. To ensure the public that such activities are being conducted with appropri-
ate controls and protections, an independent review body should be formed to
advise FDA on the ethics of the postmarketing research and surveillance activities
involving large datasets that it conducts or requires.
With regard to required RCTs, there are specific aspects of informed consent
that are more salient in the postmarketing setting than in the premarketing set -
ting. In postmarketing trials, patients may be asked to submit to a drug regimen
when a safety signal has prompted concerns about risk and possibly about the
acceptability of the drug’s benefit–risk profile. In that context, it is important to
provide information to potential participants about why a new study is required
and why it is still ethically acceptable to ask them to consider participating in
the study. Provisions may need to be made to ensure adequate discussion of how
well patients’ existing treatment is working for them. Potential participants also
need to know how the care that they will receive in an RCT may differ from the
care that they would ordinarily receive. If clinical practice shifts during the trial
period, that should be communicated to participants.
For all postmarketing research that it requires, FDA should provide the rel-
evant oversight bodies, such as the IRB and data-monitoring committee (DMC),
information about the public health question at issue; the specifics of the study
design intended to address that question, including any design features that it
views as necessary to the ethical justification of the study; and any changes in
clinical practice or professional standards that arise over the course of the study
that might affect the benefit–risk profile of a drug and influence a person’s deci -
sion to join or remain in the study. The IRB should consider dissemination of that
information to potential and current study participants.
RESPONSES TO THE CHARGE QUESTIONS9
How should FDA factor in different kinds of safety evidence in consider-
ing different kinds of regulatory actions?
Since no single algorithm can determine how to factor different kinds of
safety evidence into regulatory decision-making, the committee specifies pro -
9 The committee presents the questions in the order in which they are discussed in the chapters of
the report, not the order in which they are presented in the charge.
OCR for page 12
12 STUDYING THE SAFETY OF APPROVED DRUGS
cesses and principles to guide FDA in deciding among postmarketing regulatory
actions. The committee identifies the following five actions FDA can take to
improve its decision-making process for postmarketing regulatory decisions in
response to different kinds of safety evidence: (1) adopt a specified decision-
making framework; (2) create a BRAMP document for each drug; (3) character-
ize the nature of any disagreements about scientific evidence; (4) create effective
multidisciplinary teams with wide ranging expertise, including expertise in
observational study design and interpretation, outcomes research and pharmaco -
epidemiology, and Bayesian methods and modern causal inference approaches;
and (5) adhere to the principles of reproducible research.
What are the strengths and weaknesses of various approaches, including
observational studies, including patient registries, meta-analyses, including
patient-level data meta-analyses, and randomized controlled trials, to gener-
ate evidence about safety questions?
A wide variety of designs, data sources, and context-specific facts and priori-
ties affect the strengths and weaknesses of studies, such as whether an adverse
event is rare or common, mild or serious, or known or unknown and whether
the relative size of risk increase posed by a drug is large or small. A clinical trial
of too short a duration to find a delayed effect is going to provide less relevant
safety evidence than a design based on patient registry data with long follow-
up. These and other contextual and situation-specific factors can be expected to
trump broad principles.
The committee does, however, outline some general considerations that
are important for evaluating the value of various designs for decision-making
purposes. The initial considerations are how strong the signal that motivates
the design is and whether it involves primarily an increase in risk, a decrease
in benefit, or both, for either the general population or a definable subgroup.
Second is how time-urgent the need for a regulatory response is on the basis of
the nature of the signal. The third consideration involves how large the change
in benefits or risks must be, on both relative and absolute scales, to justify a
regulatory response. Fourth is what the other causes of a given adverse event
(or failure of benefit) might be and how predictive they are. Fifth is the qual -
ity of data likely to be gathered as part of any given design with respect to
drug exposure, outcomes, confounders, and other relevant patient, disease, or
contextual characteristics. Sixth is a judgment of how study design, conduct,
or context is likely to affect the transportability of the study results. Seventh
is what the logistical requirements of a design will be, including data access,
cost, and feasibility. Finally, there are considerations of ethical burden, consent,
confidentiality, and study oversight. These factors can lead to the choice of
either a single design type or a combination of studies with counterbalancing
strengths and weaknesses.
OCR for page 13
13
SUMMARY
Considering the speed, cost, and value of studies, what types of follow-
up studies are appropriate to investigate different kinds of signals (detected
pre-approval or postmarketing) and in what temporal order?
All the issues listed in the preceding question about design choice also affect
the optimal sequencing of study designs in response to a given signal. The main
difference is that the outcome of one study may affect the design, conduct, or ini-
tiation of subsequent studies. There are general principles to guide FDA beyond
the statutory requirement in FDAAA10 to require a clinical trial only if sufficient
information to address the public health question and the attendant decision can -
not be obtained with an observational study. All research strategies will work
best if anticipated and planned for early, even evidence synthesis. The committee
identifies a number of characteristics—consistent with FDA’s authorities under
FDAAA and with FDA guidance on postmarketing requirements—that, if evident
in the premarketing phase, should cause heightened concern about the possibility
that harm will outweigh benefit in the postmarketing context.
Under what circumstances should head-to-head randomized clinical tri-
als for safety be required?
The public health question that underlies FDA’s regulatory decisions in
the postmarketing setting is most likely to be addressed by comparing the drug
at issue with the therapies likely to be used if the drug were removed from the
market or its use were restricted. That is, it is most likely to be addressed by a
head-to-head trial involving a comparison of two active treatments that are both
indicated for the same patients who have the same condition. However, for such
a study to be scientifically valid and ethical, the active comparator must have
a well-defined benefit–risk profile and be a clinically acceptable alternative to
the drug being tested. If no comparator treatment exists or no comparator has a
well-defined benefit–risk profile, typically at least one arm of the study should
be some form of “usual care” or a placebo if usual care is not a proven or active
treatment. If there are ethical reasons for not having a usual-care or placebo arm
in the study—for example, if the treatment in question is for an irreversible and
fatal disease—a treatment that does not have a well-defined benefit–risk profile
might be the only ethically acceptable comparator. In such cases, FDA should
take the questionable benefit–risk profiles of the drug and its comparator into
account when interpreting the results of the study.
What are the ethical and informed consent issues that must be considered
when designing randomized clinical trials to evaluate potential safety risks?
The ethical justification for FDA to require a trial to resolve a postmarket -
ing benefit–risk profile question rests on the determination that (1) a responsible
10 2 1 USC § 355(o)(3)(2010).
OCR for page 18
18 STUDYING THE SAFETY OF APPROVED DRUGS
Finding 2.5
Some FDA decisions in response to postmarketing public health questions are
controversial or difficult. Complex instances tend to occur when FDA must make
a decision despite scientific disagreement about the relevant evidence or when
the likely effects of a given regulatory action are uncertain. These cases serve
as important opportunities for FDA, external scientists, and the public to learn
about the complexities of the decision-making process and the consequences of
a regulatory decision, and for FDA to improve its processes and practices.
Recommendation 2.5
FDA should conduct after-action reviews of postmarketing drug-related deci-
sions that are particularly controversial or difficult or when a major regulatory
decision is made after marketing. Such a review should include an assessment
of the decision-making process itself and the effects of the final decision on
the public’s health.
Finding 2.6
Surrogate endpoints are often relied on in the drug-approval process, and their use
has been related to a number of high-profile drug-safety problems. The findings
of postmarketing studies can be used to revise the approval process and improve
the endpoints and methods used in it.
Recommendation 2.6
As part of a continuing effort to improve regulatory science, FDA should
maintain and annually update a list of surrogate endpoints allowed for use
in the approval of drugs, the rationale for their use, the postmarketing expe -
rience regarding their correlation with health outcomes of interest, and any
revisions of approval requirements that may have been suggested by the
results of the postmarketing studies. The list should accumulate the post -
marketing experience of the successes and failures of various surrogates so
that for each major drug class, the regulatory science related to approval
methods can be modified and improved. FDA should also revise or develop
guidance documents for the use of selected surrogate endpoints that, on the
basis of postmarketing studies, appear to be inconsistently predictive of
clinical outcomes.
Finding 3.1
Some of FDA’s most difficult decisions are those in which experts disagree
about how compelling is the evidence that informs the public health question.
Understanding the nature and sources of those disagreements and their impli -
OCR for page 19
19
SUMMARY
cations for FDA’s decisions is key to improving the agency’s decision-making
process. For example, experts can disagree about the plausibility of a new risk
(or decreased benefit) on the basis of different assessments of prior evidence,
the quality of new data, the adequacy of confounding control in the relevant
studies, the transportability of results, the appropriateness of the statistical
analysis, the relevance of the new evidence to the public health question, how
the evidence should be weighed and synthesized, or the threshold for regula -
tory actions.
Recommendation 3.1
FDA should use the framework for decision-making proposed in Recom-
mendation 2.1 to ensure a thorough discussion and clear understanding of the
sources of disagreement about the available evidence among all participants
in the regulatory decision-making process. In the interest of transparency,
FDA should use the BRAMP document proposed in Recommendation 2.2 to
ensure that such disagreements and how they were resolved are documented
and made public.
Finding 3.2
Such methods as Bayesian analyses or other approaches to integrating external
relevant information with newly emerging information could provide decision-
makers with useful quantitative assessments of evidence. An example would be
sensitivity analyses of clinical-trial data that illustrate the influence of prior prob -
abilities on estimates of probabilities that an intervention has unacceptable safety
risks. These approaches can inform judgments, allow more rational decision-
making, and permit input from multiple stakeholders and experts.
Recommendation 3.2
FDA should ensure that it has adequate expertise in Bayesian approaches, in
combination with expertise in relevant frequentist and causal inference meth -
ods, to assess the probability that observed associations reflect actual causal
effects, to incorporate multiple sources of uncertainty into the decision-
making process, and to evaluate the sensitivity of those conclusions to dif -
ferent representations of external evidence. To facilitate the use of Bayesian
approaches, FDA should develop a guidance document for the use of Bayes -
ian methods for assessing a drug’s benefits, risks, and benefit–risk profile.
Finding 3.3
Traditionally, the main criteria for evaluating a study are ones that contribute to
its internal validity. A well-conducted RCT typically has higher internal valid -
ity than a well-conducted observational study. Results of observational studies,
OCR for page 20
20 STUDYING THE SAFETY OF APPROVED DRUGS
however, can have greater transportability if their participants are more similar
to the target clinical population than to the participants in a clinical trial. In some
circumstances, such as an evaluation of the association between a drug and an
uncommon unexpected adverse event, observational studies may produce esti -
mates closer to the actual risk in the general population than can be achieved in
clinical trials. In assessing the relevance of study findings to a public health ques-
tion, the transportability of the study results is as important as the determinants
of its internal validity.
Recommendation 3.3
In assessing the benefits and risks associated with a drug in the postmarketing
context, FDA should develop guidance and review processes that ensure that
observational studies with high internal validity are given appropriate weight
in the evaluation of drug harms and that transportability is given emphasis
similar to that given bias and other errors in assessing the weight of evidence
that a study provides to inform a public health question.
Finding 3.4
The principles of reproducible research are important for ensuring the integrity of
postmarketing research used by FDA. Those principles include providing informa-
tion on the provenance of data (from measurement to analytic dataset) and, when
possible, making available properly annotated analytic datasets, study protocols
(including statistical analysis plan) and their amendments, and statistical codes.
Recommendation 3.4
All analyses, whether conducted independently of FDA or by FDA staff,
whose results are relied on for postmarketing regulatory decisions should use
the principles of reproducible research when possible, subject to legal con-
straints. To that end, FDA should present data and analyses in a fashion that
allows independent analysts either to reproduce the findings or to understand
how FDA generated the results in sufficient detail to understand the strengths,
weaknesses, and assumptions of the relevant analyses.
Finding 3.5
The ability of researchers in and outside FDA to analyze new information about
the benefits and risks associated with a marketed drug and to design appropriate
postmarketing research—including conducting individual-patient meta-analy -
ses—is enhanced by access to data and analyses from all studies of the drug
and others in the same drug class that were reported in the preapproval process.
Although disclosure of such information is likely to advance the public’s health,
such disclosures raise concerns about the privacy of participants in the research
OCR for page 21
21
SUMMARY
that generated the information and may threaten industry interest in maintain -
ing proprietary information, which is deemed important for innovation. New
approaches to resolving this tension are needed.
Recommendation 3.5
FDA should establish and coordinate a working group, including industry and
patient and consumer representatives, to find ways that appropriately balance
public health, privacy, and proprietary interests to facilitate disclosure of data
for trials and studies relevant to postmarketing research decisions.
Finding 3.6
The elements of the benefit–risk profile of a drug are best estimated by using all
the available high-quality data, and meta-analysis is a useful tool for summarizing
such data and evaluating heterogeneity. However, because the reporting of harms
in published RCTs and observational studies is often poor or inconsistent and
because there is often substantial publication bias in studies of drug risk, steps
are needed to improve both the reporting of harms and the design of studies of
harm. That can be done through prospective planning for selected meta-analyses
and by monitoring compliance with the FDAAA requirement that summary trial
results for all primary and secondary outcomes be published at ClinicalTrials.gov.
Recommendation 3.6
For drugs that are likely to have required postmarketing observational stud -
ies or trials, FDA should use the BRAMP document to specify potential
public health questions of interest as early as possible; should prospectively
recommend standards for uniform definition of key variables and complete
ascertainment of events among studies or convene researchers in the field to
suggest such standards and promote data-sharing; should prospectively plan
meta-analyses of the data with reference to specified exposures, outcomes,
comparators, and covariates; should conduct the meta-analyses of the data;
and should make appropriate regulatory decisions in a timely fashion. FDA
can also improve the validity of meta-analyses by monitoring and encourag -
ing compliance with FDAAA requirements for reporting to ClinicalTrials.
gov.
Finding 3.7
FDA produced a high-quality guidance document on the use of the noninferiority
design for the study of efficacy. Increasingly, FDA is using the noninferiority design
to evaluate drug safety endpoints as the primary outcomes in randomized trials.
The use of noninferiority analyses to establish the acceptability of the benefit–risk
profile of a drug can take the decision about how to balance the risks and benefits
OCR for page 22
22 STUDYING THE SAFETY OF APPROVED DRUGS
of two drugs out of the hands of regulators. Noninferiority trials also have the
disadvantage of being biased toward equivalence when trial design or conduct is
suboptimal; this is of particular concern when such trials are used to estimate risks.
Recommendation 3.7.1
FDA should develop a guidance document on the design and conduct of
noninferiority postmarketing trials for the study of safety of a drug. The guid-
ance should include discussion of criteria for choosing the standard therapy
to be used in the active-treatment control arm; of methods for selecting a
noninferiority margin in safety trials and ensuring high-quality trial conduct;
of the optimal analytic methods, including Bayesian approaches; and of the
interpretation of the findings in terms of the drug’s benefit–risk profile.
Recommendation 3.7.2
FDA should closely scrutinize the design and conduct of any noninferiority
safety studies for aspects that may inappropriately make the arms appear
similar. FDA should use the observed-effect estimate and confidence interval
as a basis for decision-making, not the binary noninferiority verdict.
Finding 4.1
A decision by FDA to require postmarketing research can put research participants
at risk. It can also put patients and the public at risk by delaying a regulatory deci-
sion that might be protective of public health. Some conditions are necessary but
not sufficient for an FDA decision to require postmarketing research to be ethical.
Recommendation 4.1.1
FDA should require postmarketing research only when (1) uncertainty about
the benefit–risk balance is such that a responsible decision about the future
regulatory status of the drug cannot be made on the basis of existing evidence;
(2) it is expected that the research can be properly designed and implemented
to reduce uncertainty about the benefit–risk profile to allow a responsible
regulatory decision; (3) FDA has a plan for using the results of the research
to make a regulatory decision in a timely fashion; and (4) the research can
be conducted in a manner that provides sufficient protection of and respect
for research participants.
Finding 4.2
For postmarketing investigations authorized under Section 901 of FDAAA,12
FDA can require an RCT only if it is unable to obtain the data that it needs from
an observational study or surveillance. Determining what kind of study will pro -
12 21 USC § 355(o) (2010).
OCR for page 23
23
SUMMARY
vide the information needed to answer FDA’s public health question, however, is
complex. In the postmarketing setting, both observational studies and RCTs have
advantages and disadvantages. In some circumstances, the evidence provided by
an observational study may be as good as or better for informing a public health
question than the evidence provided by a feasible clinical trial; that is more likely
to occur when the magnitude of the relative risk is large in contrast with the
potential for confounding, which occurs with many drug harms. Observational
studies also have a number of ethical and practical advantages over RCTs. In
other circumstances, however, the evidence available from an observational study
would not be able to provide the necessary additional information to help answer
the public health question. Those instances are more likely to occur when the
public health questions are related primarily to a drug’s benefits.
Recommendation 4.2
When deciding which type of research to require in the postmarketing setting,
FDA should carefully weigh the strengths of potential observational studies
for evaluating risks and their ethical and practical advantages, including the
timeframe within which the data are needed, against the limitations of poten -
tial observational studies for generating the data needed to answer the public
health question. An RCT should be required only if FDA has concluded that
an observational study could not provide the necessary information, that an
RCT is likely to generate the information within the necessary timeframe,
and that the necessary RCT is ethically acceptable.
Finding 4.3
When FDA requires a postmarketing RCT, the public health question is most likely
to be properly addressed by a comparison of the target drug with the standard ther-
apy for the condition involved—if there is a standard therapy. Such a trial would
involve a “head-to-head” design, defined as a comparison of two active treatments
that are indicated for the same patients who have the same condition. However, it is
also important both scientifically and ethically for at least one clinically acceptable
comparator in the required trial to have a well-defined benefit–risk profile.
Recommendation 4.3
If FDA requires a postmarketing RCT for an indication for which there is an
accepted active treatment that would probably be used if access to the drug
under study were restricted, the alternative treatment should be used as at
least one comparator in the trial.
Finding 4.4
When deciding whether to require a postmarketing study, FDA must balance
its ethical obligation to protect the public’s health with its ethical obligation to
OCR for page 24
24 STUDYING THE SAFETY OF APPROVED DRUGS
protect research participants. In some instances, FDA may be faced with a deci -
sion to require an RCT that might expose participants to more net risk than they
would probably face if decisions about their drug treatment were being made in
the context of clinical practice or that offers no reasonable expectation of clinical
benefit to participants although its results may benefit society. Requiring such a
study may be ethically justifiable but only under special circumstances.
Recommendation 4.4
FDA should require a postmarketing RCT that might expose research par-
ticipants to more risk or less net clinical benefit than they would probably
face if decisions about their drug treatment were being made in the context
of clinical practice only if a question of pressing public health significance is
at stake, if no other design with a better benefit–risk balance for participants
could supply the evidence needed for a responsible regulatory response to
the question, and if special safeguards are in place to protect the rights and
interests of the research participants. Those safeguards should include the
determination by an appropriately constituted review committee that the
additional risk is small enough for it to be ethical to ask people whether they
are willing to accept it solely to contribute to the public good; the minimiza-
tion of additional risk by careful study design and implementation of a robust
monitoring plan throughout the study; the inclusion of special measures in
the process of soliciting informed consent to confirm that patients understand
and willingly accept that they are assuming an additional risk, beyond what
they are likely to face in clinical practice, solely in the interest of the public
good; and the implementation of processes to ensure that over the course
of the trial participants are regularly informed of any changes in clinical
practice or the medical literature relevant to assessments of the comparative
benefits and risks associated with trial participation and (nonresearch) clini -
cal management.
Finding 4.5
Although regulations governing human subjects research do not apply if an activ -
ity is considered public health practice, as is the case with the Sentinel system,
it is often not possible to draw a clear or ethically relevant distinction between
some kinds of FDA-required observational research and public health practice.
It is important that FDA, in conjunction with the Office for Human Research
Protections (OHRP), clarify whether its human subjects regulations (21 CFR 50)
govern required postmarketing observational studies and, if so, how FDA will
address and will expect IRBs to address any differences between 21 CFR 50 and
other potentially applicable human subject regulations (45 CFR 46 Subpart A)
in oversight and research-participant protection, including consent requirements,
OCR for page 25
25
SUMMARY
in different observational designs so that its regulations are not a barrier to what
would otherwise be ethically acceptable observational designs. FDA also needs
to determine how best to ensure that it is feasible for drug companies and their
contractors to conduct the postmarketing observational studies that it requires,
in view of the Health Insurance Portability and Accountability Act of 1996 and
other potential constraints, while protecting the privacy of the people whose data
are used. It is also likely that the desirability of linking datasets and of obtaining
additional information from patients or otherwise needing access to some iden -
tifying information about patients will increase, whether studies are conducted
under the auspices of FDA-supported surveillance systems, such as Sentinel and
deemed public health practice, or conducted by manufacturers as required by
FDA and interpreted at least by some to be research, raising additional ethical
questions about the adequacy of data security, authorization of access to different
datasets, and different research and public health purposes.
Recommendation 4.5.1
FDA, in conjunction with the Office for Human Research Protections
(OHRP), should clarify whether its human subjects regulations (21 CFR 50)
govern required postmarketing observational studies and, if so, how FDA
will address and will expect IRBs to address any differences between 21
CFR 50 and other potentially applicable human subject regulations (45 CFR
46 Subpart A) in oversight and research-participant protection, including
consent requirements.
Recommendation 4.5.2
To assure the public that surveillance and required observational studies can
proceed with appropriate controls and protections, and to facilitate the con -
duct of ethically acceptable surveillance and required observational studies
that are important to the public’s health, FDA should form an independent
body to advise FDA, on an as needed basis, on the ethics of postmarketing
research and surveillance activities that it conducts or requires. This advisory
body should be positioned to provide guidance on emerging ethical chal -
lenges, with particular focus on activities that are determined not to require
IRB oversight.
Finding 4.6
FDA has an ethical obligation to ensure that the rights and interests of participants
in the postmarketing research that it requires are properly protected. IRBs and
data-monitoring committees (DMCs) can play a critical role in assisting FDA
with this obligation, but these bodies require information and guidance from FDA
to be effective in their research-participant protection responsibilities.
OCR for page 26
26 STUDYING THE SAFETY OF APPROVED DRUGS
Recommendation 4.6
For all postmarketing research that it requires and that is subject to IRB or
DMC oversight, FDA should provide each IRB (including centralized IRBs
and multiple IRBs) and each DMC with the up-to-date BRAMP document
for the study drug and sufficient information in writing for the IRB or DMC
to provide appropriate oversight, including information about the public
health question at issue, the specifics of the study design intended to address
the question, design features that FDA views as necessary for the ethical
justification of the study, and any changes in clinical practice or profes-
sional standards that arise over the course of the study that might affect the
benefit–risk profile of a drug and influence a person’s decision to participate
or remain a participant in the study.
Finding 4.7
There are heightened informed consent concerns in the conduct of FDA-required
RCTs in the postmarketing setting. FDA has an ethical responsibility to ensure
that postmarketing clinical trials include appropriate informed consent processes
and oversight.
Recommendation 4.7
FDA should issue guidance for interpreting disclosure and informed consent
requirements in applicable federal regulations in the context of postmarket -
ing RCTs that it requires, using the authorities granted to it in Section 901 of
FDAAA13 to help oversight bodies (such as IRBs) to ensure that such trials
include a comprehensive informed consent process. The guidance should
emphasize that, in addition to standard disclosure requirements, the follow -
ing information of particular importance in the postmarketing setting should
be communicated to research participants: why a new study of an approved
drug is being required; salient risks posed by participation in required post -
marketing research, including whether new information suggests that the drug
under study may pose serious risks; and whether medical practice has shifted
or is shifting away from prescribing the study drug. The guidance should
make clear that participants must be informed of any substantial changes in
clinical practice and professional standards over the course of the trial and
informed of any new research findings relevant to their willingness to accept
or to continue to accept the risks associated with the trial. And the guidance
should identify the conditions under which consent processes should include
measures to validate the adequacy of participants’ understanding, not only
the adequacy of the disclosures made to participants.
13 2 1 USC § 355(o) (2010).
OCR for page 27
27
SUMMARY
Finding 4.8
During the last two decades, the volume of clinical trials conducted outside the
United States has increased dramatically, and this has led to concerns about the
quality, reliability, and transportability of research results and about the adequacy
of protections for research participants. Those concerns apply as well to FDA-
required postmarketing research that uses research sites outside the United
States. FDA’s Office of International Programs, through its Harmonization and
Multilateral Relations Office, is tasked with the responsibility of coordinating and
collaborating with other agencies and countries on international standards and
harmonization issues and is therefore well positioned to address these concerns.
Recommendation 4.8
FDA should direct its Office of International Programs to include explicitly
among its responsibilities working with counterpart agencies of other gov-
ernments and with industry to resolve concerns about the ethics and quality
of evidence in the conduct of FDA-required postmarketing research outside
the United States.
OCR for page 28
