Reproductive and Developmental Toxicities -- Integrating Study Results to Assess Concerns

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

Guidance for Industry
Reproductive and
Developmental Toxicities —
Integrating Study Results to
Assess Concerns
U.S. Department of Health and Human Services
Food and Drug Administration
Center for Drug Evaluation and Research (CDER)
September 2011
Pharmacology and Toxicology

Guidance for Industry
Reproductive and
Developmental Toxicities —
Integrating Study Results to
Assess Concerns
Additional copies are available from:
Office of Communications, Division of Drug Information
Center for Drug Evaluation and Research
Food and Drug Administration
10903 New Hampshire Ave., Bldg. 51, rm. 2201
Silver Spring, MD 20993-0002
Tel: 301-796-3400; Fax: 301-847-8714; E-mail: druginfo@fda.hhs.gov
http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/default.htm
U.S. Department of Health and Human Services
Food and Drug Administration
Center for Drug Evaluation and Research (CDER)
September 2011
Pharmacology and Toxicology

TABLE OF CONTENTS
I. 	INTRODUCTION............................................................................................................. 1
II. 	BACKGROUND ............................................................................................................... 2
A. 	Data Needed for Integration and Assessment ............................................................................. 2
B. 	Types of Reproductive and Developmental Toxicity Evaluated................................................ 3
1. Reproductive Toxicities.................................................................................................................... 3
2. Developmental Toxicities................................................................................................................. 4
III. 	THE INTEGRATION AND ASSESSMENT PROCESS.............................................. 5
A. 	Overall Decision Tree (Figure A) ................................................................................................. 5
1. Availability of Studies ...................................................................................................................... 6
2. Relevance of Studies ........................................................................................................................ 6
3. Presence or Absence of a Signal...................................................................................................... 6
B. 	No Signal (Figure B) ...................................................................................................................... 7
1. The Model/Test Species Predictive Adequacy ................................................................................. 7
2. Adequacy of Study Doses and Exposure .......................................................................................... 8
3. Class Alert........................................................................................................................................ 8
4. Signals for Related Types of Reproductive and Developmental Toxicity ........................................ 8
C. 	Reproductive or Developmental Toxicity Endpoints with Positive Signal (Figure C) ............ 9
1. Overview of Integration ................................................................................................................... 9
2. Factors ........................................................................................................................................... 10
3. Summary/Integration of Positive Findings .................................................................................... 14
GLOSSARY................................................................................................................................. 16
FIGURES..................................................................................................................................... 18

Contains Nonbinding Recommendations
1
Guidance for Industry 1
Reproductive and Developmental Toxicities — Integrating Study
Results to Assess Concerns
This guidance represents the Food and Drug Administration’s (FDA’s) current thinking on this topic. It
does not create or confer any rights for or on any person and does not operate to bind FDA or the public.
You can use an alternative approach if the approach satisfies the requirements of the applicable statutes
and regulations. If you want to discuss an alternative approach, contact the FDA staff responsible for
implementing this guidance. If you cannot identify the appropriate FDA staff, call the appropriate
number listed on the title page of this guidance.
I. 	INTRODUCTION
This guidance describes an approach to estimating possible human developmental or
reproductive risks associated with drug or biological product exposure when a nonclinical
finding of toxicity has been identified, but definitive human data are unavailable. The guidance
is intended for applicants of new drug applications (NDAs) and biologics licensing applications
(BLAs). The recommendations included here will also help to ensure a consistent review of
reproductive and developmental toxicity data among Center for Drug Evaluation and Research
(CDER) review staff.
This guidance does not (1) give detailed advice about labeling or placement of toxicity
information in product labeling (for information on labeling, see 21 CFR 201.57); or (2) discuss
clinical data or the integration of nonclinical and clinical data.
FDA guidance documents, including this one, 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.
1 This guidance was developed in the Center for Drug Evaluation and Research at the Food and Drug
Administration.

Contains Nonbinding Recommendations
2
II. 	BACKGROUND
The approach presented here for assessing nonclinical reproductive and developmental toxicity
data involves the integration and careful consideration of a variety of different types of
nonclinical information: reproductive toxicology; general toxicology; and toxicokinetic and
pharmacokinetic information, including absorption, distribution, metabolism, and elimination
data. The approach is used when there is a toxicity finding and focuses on assessing the
likelihood that a drug will increase the risk of adverse human developmental or reproductive
outcomes. The approach includes noting when studies were not conducted or when they were not
performed using relevant model systems or at appropriate dose ranges.
The general principles described here (i.e., a comprehensive analysis of available data) will
typically be relevant to both drug and biological products, although some factors may not apply
to biological products, because data may not be available for all factors considered in this
guidance (e.g., cross-species concordance, dose-response, metabolism, relative exposure (animal
: human) of >25-fold). For some oncology products (e.g., cytotoxics), certain aspects of the
guidance may not apply because patients may be dosed at the maximum tolerated dose (MTD).
Note: Available clinical information to evaluate a drug’s potential to increase the risk of an
adverse developmental or reproductive outcome in humans should be evaluated separately and,
when definitive, can supersede any nonclinical findings.
A. 	Data Needed for Integration and Assessment
When determining what nonclinical data will be needed for integration and assessment, it is
important to first evaluate a complete set of the expected general toxicology, reproductive and
developmental toxicology, and pharmacokinetic studies.2 The evaluation should include an
assessment of the drug’s ability to produce a positive finding in the relevant animal studies (e.g.,
whether doses used were high enough to induce toxicity of some kind). The evaluation should
also compare animal and human pharmacodynamic effects, animal and human metabolism and
disposition, animal and human pharmacologic and toxic effects, and drug exposures in animal
studies in relation to the highest proposed dose in humans. The type and extent of available
toxicology data may vary, depending on the product’s biological actions, test systems available
for studying the compound, and other factors. In some cases, the data will not include all
desirable general toxicology, reproductive/developmental toxicology, and pharmacokinetic
studies. Nevertheless, the product should be evaluated to the extent possible according to the
scientific principles and considerations described in this document (see section III).
2 See the following International Conference on Harmonisation (ICH) guidances for industry: M3(R2) Nonclinical
Safety Studies for the Conduct of Human Clinical Trials for Pharmaceuticals; S3A Toxicokinetics: The Assessment
of Systemic Exposure in Toxicity Studies; S5(R2) Detection of Toxicity to Reproduction for Medicinal Products &
Toxicity to Male Fertility; and S9 Nonclinical Evaluation for Anticancer Pharmaceuticals.

Contains Nonbinding Recommendations
3
B. 	Types of Reproductive and Developmental Toxicity Evaluated
Reproductive and developmental toxicity can be manifested as a change in one or more specific
endpoints. For the purposes of this document, there are two broad toxicity categories —
reproductive and developmental — and, within those categories, eight classes of effects are
considered (see Glossary).
Classes of reproductive toxicity include:
– 	male fertility
– 	female fertility
– 	parturition
– 	lactation.
Classes of developmental toxicity include:
– 	mortality
– 	dysmorphogenesis (structural abnormalities)
– 	alterations to growth
– 	functional impairment.
For a given drug, endpoints reflecting the full range of potential reproductive and developmental
effects should ordinarily be assessed (for standard endpoints, see ICH S5(R2)). A positive signal
of reproductive or developmental toxicity, whether in valid reproductive/developmental or other
relevant nonclinical studies, should be evaluated to estimate the likelihood of increased
reproductive or developmental risk for humans.
1. 	Reproductive Toxicities
Reproductive toxicity refers to structural and functional alterations that affect reproductive
competence in sexually mature males and females. Reproductive toxicity includes effects on
male fertility and female fertility, parturition, and lactation.
• 	Male Fertility
Male reproductive toxicity includes damage to the reproductive organs, alterations in endocrine
regulation of gamete maturation and release, reduction in sperm count, alterations in sperm
motility or morphology, aberrant mating behavior, altered ability to mate, alterations in
endocrine function, or overall reduction in fertility.
• 	Female Fertility
Female reproductive toxicity includes damage to the reproductive organs, alterations in
endocrine regulation of gamete maturation and release, aberrant mating behavior, altered ability
to mate, or overall reduction in fertility. Diminished fertility in female animals is typically

Contains Nonbinding Recommendations
4
detected by reductions in the fertility index, the number of implantation sites, time to mating, or
fecundity.
• 	Parturition
Toxicities affecting labor and delivery in animals include abnormal or difficult delivery
(dystocia) and changes in the onset or duration of parturition. Changes in the duration of
parturition are frequently reported as mean time elapsed per pup, or total duration of parturition.
• 	Lactation
Drugs may alter the process of lactation in the nursing mother (e.g., the quality or quantity of
milk), or may alter maternal behavior towards the nursing offspring.
2. 	Developmental Toxicities
Developmental toxicity refers to adverse effects on the developing organism that result from
exposure prior to conception, during the prenatal period, or postnatally up to the time of sexual
maturity. The four major manifestations of developmental toxicity are:
– 	mortality
– 	dysmorphogenesis (structural abnormalities)
– 	alterations to growth
– 	functional impairment
• 	Mortality
Mortality resulting from developmental toxicity may occur at any time from early conception to
post-weaning (e.g., embryo-fetal death is a subset of mortality resulting from developmental
toxicity). Thus, a positive signal may appear as:
– 	pre- or peri-implantation loss
– 	early or late resorption
– 	abortion
– 	stillbirth
– 	neonatal death
– 	peri-weaning loss
• 	Dysmorphogenesis (Structural abnormalities)
Dysmorphogenic effects are generally seen as malformations or variations of the skeleton or soft
tissues of the offspring and are commonly referred to as structural abnormalities.

Contains Nonbinding Recommendations
5
• 	Alterations in Growth
Alterations in growth are generally seen as growth retardation, although excessive growth or
early maturation may also be considered alterations to growth. Body weight is the most common
measurement for assessing growth rate. Crown-rump length and anogenital distance may also be
measured. Sometimes it is not clear if an effect is a direct structural alteration or an inhibition of
growth. For example, reduced ossification could be either. A distinction must be made upon
review of all relevant data.
• 	Functional Impairment
Functional toxicities could include any persistent alteration of normal physiologic or biochemical
function, but typically only developmental neurobehavioral effects and reproductive function are
measured. Common assessments include:
– 	locomotor activity
– 	learning and memory
– 	reflex development
– 	time to sexual maturation
– 	mating behavior
– 	fertility
III. 	THE INTEGRATION AND ASSESSMENT PROCESS
Recommendations for wording in labeling should be based on the results of the integration and
assessment process. The specific considerations leading to a risk conclusion should be provided.
This information may later be helpful in discussions between FDA reviewers and applicants.
The complete data integration process is divided into three components as discussed below in
Sections A through C and presented schematically in Figures A through C at the end of this
document. Briefly, Figure A is applicable to all data sets, Figure B is applicable only to data sets
without evidence of reproductive or developmental toxicity, and Figure C is applicable to data
sets with positive indications of reproductive or developmental toxicity.
A. 	Overall Decision Tree (Figure A)
The decision process outlined in Figure A should be used for any of the endpoints of
reproductive or developmental toxicity discussed in Section II.B. For a given drug, studies may
have been conducted to evaluate potential effects on none, some, or all of the endpoints of
reproductive and developmental toxicity. Where studies are available for any of the different
endpoints, the outcome may be one or more positive signals, or no positive signal. It is
recognized that, in practice, one study may address several endpoints of toxicity and a study may
be considered adequate to evaluate all, some, or none of the endpoints of toxicity addressed.
Figure A depicts the sequential decisions that should be made in evaluating the various situations

Contains Nonbinding Recommendations
6
that may be encountered and the next steps that should be taken when there are evaluable studies
with positive or negative findings.
1. 	Availability of Studies
In Figure A, the first question asked for each category or class of toxicity is: "Were studies
performed to assess the risk of that type of toxicity in humans, and are the detailed study results
available for comprehensive evaluation?" If no studies were conducted, or detailed study results
are unavailable for comprehensive evaluation, the NDA or BLA review or applicant submission
should explain that studies to assess the risk of that type of toxicity were not done, or are
otherwise unavailable. In such circumstances, risk to humans is considered unknown or not
evaluable.
Example of wording: The risk of human [reproductive or developmental toxicity] with
[Drug X] is unknown. There are no or inadequate data to evaluate its potential to cause
human [reproductive or developmental toxicity].
If studies were conducted and are available for comprehensive evaluation, the assessment
process should continue with question 2.
2. 	Relevance of Studies
The next question asked for each category or class of toxicity is: "Do the studies provide
information relevant to assessing the risk of that type of toxicity for the proposed human use?"
If the studies were not relevant or were otherwise inappropriate (i.e., inappropriate test protocol,
nonrelevant route of drug administration), the NDA or BLA review or applicant submission
should explain why and should discuss all supporting information that bears on study relevance.
For example, the studies might have been potentially relevant, but the study design or conduct
may have resulted in insufficient information to be useful. If the study was not relevant, the risk
to humans is considered unknown or not evaluable.
Example of wording: Animal data are insufficient to assess the risk of human
[reproductive or developmental toxicity] with [Drug X].
If the studies conducted were relevant to evaluating the risk of the particular type of toxicity in
humans, the risk integration process should continue with question 3. Note that the processes in
Figures B and C (see end of document) are intended to be used only when studies are considered
adequate to assess the specified risk. They should not be used to evaluate findings (positive or
negative) derived from inadequate studies.
3. 	Presence or Absence of a Signal
If relevant studies are available and the study conduct (including doses and exposure and route
of administration) is appropriate for assessing the risk of toxicity in humans, the next question
asked should be, “Was there a positive signal (suggesting toxicity)?” If no signal was seen, the
evaluation process should continue per Section B (Figure B). A positive signal is a biologically

Contains Nonbinding Recommendations
7
meaningful difference in dosed animals compared to concurrent or historical controls. If a
positive signal was seen, the evaluation process should continue per Section C (Figure C).
If multiple studies are available to assess a particular type of reproductive or developmental
toxicity (e.g., multiple studies would be expected for the evaluation of effects on embryonic
development – ICH stage C), the process in Figure B should be used only if the results of all
studies relevant to a particular aspect of reproductive or developmental toxicity were negative
for that type of toxicity.
If any study (general toxicity, reproductive, or developmental toxicology study) had a positive
signal for that aspect of reproductive or developmental toxicity, the process in Section C (Figure
C) should be used.
B. 	No Signal (Figure B)
When there is no positive signal for an endpoint of reproductive or developmental toxicity, the
risk assessment should be a step-wise process leading to a discussion in the NDA or BLA review
or applicant submission of the applicability of the non-finding to humans. A graphic
representation of this process is presented in Figure B.
The following four sets of questions should be considered during the evaluation of each type of
reproductive or developmental toxicity for which there was no signal.
1. 	The Model/Test Species Predictive Adequacy
To what extent are the models or test species used likely to be predictive of human response?
The questions below bear on determining a model’s predictive adequacy. Affirmative answers
to these questions would make the findings in that test system more credible in terms of human
relevance.
Do the models or test species (or systems) demonstrate the intended pharmacodynamic effect(s)
of the drug?
Do the models/test species (or systems) demonstrate an overall toxicity profile that is consistent
with the human toxicity profile?
Do the models/test species (or systems) demonstrate pharmacokinetic (including ADME)
profiles for the drug that are qualitatively similar to those in humans?
Negative answers to these questions may suggest that the response of the test species is of little
relevance to humans, and the BLA or NDA review or applicant submission should explain why
the animal study or studies conducted with the drug may not be fully adequate to evaluate the
risk for the particular type of toxicity in humans (i.e., why the test may have low predictive
value). Even if the test system is determined to be of limited relevance based on the above
considerations, the review or applicant submission should consider the remaining questions (2–4

Contains Nonbinding Recommendations
8
below) and describe any additional uncertainties associated with the nonclinical data and the
relevance of the studies to humans.
2. 	Adequacy of Study Doses and Exposure
The following elements should also be considered in assessing the relevance to humans of the
drug doses and exposure in the test system:
Were adequate doses (concentrations) of the drug administered to the test species or test systems
(e.g., MTD or maximum feasible dose (MFD); see ICH-S5R2 Note 7 (3.1))? (This would usually
not apply to biologics.)
Were the drug exposures (based on AUC, Cmax , or other appropriate systemic exposure metric)
achieved in the test species or test systems adequate relative to those expected in humans at the
maximum recommended human dose (generally some multiple of the human exposure but at
least equivalent to it)? A greater relative exposure adds to the credibility of a negative finding.
If the answer to either of these questions is no, the review or applicant submission should state
that the animal studies conducted may be inadequate to fully evaluate the risk for the particular
type of toxicity reported to be negative and should explain in detail why they may be inadequate.
Even if the study doses and exposure are considered inadequate, the evaluation should proceed
to the remaining sections (3–4 below), and any additional uncertainties should be described in
the evaluation.
3. 	Class Alert
Is there a class alert? Class alerts should be based on adverse reproductive or developmental
effects previously demonstrated in humans by closely related chemical entities or compounds
with similar pharmacodynamic effects. If there is a class alert for the drug based on a related
chemical structure of parent or metabolite or related pharmacologic effect, the class-specific
information relevant to the type of toxicity reported to be negative should be included in the risk
evaluation and discussion of the drug.
4. 	Signals for Related Types of Reproductive and Developmental Toxicity
The next step in evaluating the significance of a no-signal finding for a particular type of
reproductive or developmental toxicity is to consider whether or not there are findings for related
reproductive and developmental toxicities. A positive signal for one endpoint of toxicity may
suggest some risk in humans for other toxicities in the same category for which there were no
findings in animals. This may be a consequence of limitations of studies or cross-species
differences in expression of effects.
The issue of related toxicities is most applicable to developmental toxicities. For example, if
there is no signal for fetal mortality, but a positive signal for alterations to growth or
dysmorphogenesis in one (or more) animal species, it may be inappropriate to conclude there is
no risk of fetal mortality for humans. Related toxicities may also be pertinent for reproductive

Contains Nonbinding Recommendations
9
toxicities where a hormonal mechanism is identified that could impact multiple aspects of
reproductive performance and is relevant to humans. If positive signals for related endpoints of
toxicity were observed in the animal studies, the evaluation should state that there was no
observed effect on the type of toxicity being assessed, but positive signals were seen for related
toxicities. If there is no positive signal for any type of reproductive or developmental toxicity in
adequate studies, the evaluation should state that there is no expected increase in risk for
reproductive or developmental toxicity in humans, based on the results of animal studies.
C. 	Reproductive or Developmental Toxicity Endpoints with Positive Signal (Figure C)
1. 	Overview of Integration
A positive nonclinical signal for any type of reproductive or developmental toxicity should be
analyzed with respect to various factors that may affect the level of concern for adverse effects in
humans. Since multiple factors contribute to the overall evaluation and conclusion, scientific
judgment should be used to integrate all of the factors applicable to positive findings.
These factors, discussed in greater detail below, include the following.
– 	Cross-species concordance of reproductive or developmental effects
– 	Multiplicity of effects
– 	Maternal/paternal toxicity
– 	Dose–response relationship
– 	Rare events
– 	Pharmacodynamics: similarity between pharmacologic and toxicologic mechanisms
– 	Concordance between test species and humans: metabolic and general toxicity profiles
– 	Relative exposure
– 	Class alerts
These and other factors can either increase or decrease the level of concern. Examples of factors
increasing concern when associated with a given positive signal include: low relative exposure in
animals, presence of cross-species concordance, absence of maternal toxicity, and similarity
between pharmacologic and reproductive/developmental toxicologic mechanisms. Conversely,
concern may be decreased by: high relative exposure in animals, absence of cross-species
concordance, excessive maternal toxicity, and animal-specific mechanisms. Case by case, some
factors, such as relative exposure, can have a greater impact than others. Some of the factors
may not apply to biologics or to products for oncology indications because the relevant data
(e.g., from multiple species) may not exist or be expected (see ICH S6(R1)). The review or
applicant submission should discuss the various factors and describe how the overall conclusion
was reached.
The implicit assumption of the integrative assessment shown in Figure C is that the process
begins with a positive signal that is evident in one or more of the examined species (either in a
reproductive or developmental toxicology study or an effect on a reproductive
tissue/system/behavior in a general toxicology study).

Contains Nonbinding Recommendations
10
Note: Human data are considered separately from nonclinical findings, and human data may
dramatically influence the overall assessment of human risk of reproductive or developmental
toxicity, since definitive human data would supersede any nonclinical data.
2. 	Factors
• 	Cross-species concordance of reproductive or developmental effects
The observation of positive signals for the same or a related type of toxicity in more than one
species is described as cross-species or interspecies concordance. In general, findings for which
there is cross-species concordance are more convincing than findings in a single species; thus,
there is increased concern for reproductive or developmental toxicity in humans when cross-
species concordance of these effects is observed in nonclinical studies. In evaluating potential
human risk for adverse reproductive or developmental outcomes, if there is cross-species
concordance for a single type of adverse effect it may be reasonable to conclude that a similar
effect is the most likely adverse event to be seen in humans treated with the drug. However,
cross-species concordance is not limited to the observation of the same specific effect across
species. If different but related adverse effects are seen in multiple test species (e.g., alterations
to growth in one species and developmental mortality in another, or parturition effects in one
species and lactation effects in the second), it is assumed that there is an increased level of risk
for categorically related endpoints in humans.
Cross-species concordance is most likely to be identified for developmental endpoints examined
in the organogenesis testing paradigm, in which multiple species are typically evaluated.
However, alterations to endocrine function or gonadal histopathology (which may alter fertility)
may be indirectly detected in subchronic and chronic toxicity studies in rodents and nonrodents.
For alterations to parturition or lactation, it is often not possible to assess cross-species
concordance because pre- and postnatal studies to assess these toxicity classes are usually done
in only a single species. Although there may be less concern when a signal is detected in only
one species, it is important to determine whether the negative species is an appropriate animal
model and the studies were adequate in design, dosing, and implementation.
• 	Multiplicity of effects
Multiplicity of effects refers to the observation, in a single species or animal model, of two or
more positive signals within one of the two general categories of toxicity (reproductive or
developmental) or within one of the classes of reproductive or developmental toxicity. Evidence
of multiple effects in an animal species is associated with increased concern for reproductive or
developmental toxicity in humans, while an isolated finding is generally of less concern. The
observation of positive signals in multiple classes of reproductive or developmental toxicity
(e.g., increased malformation and embryo-fetal death) or of multiple signals for the same class of
effects (e.g., motor and cognitive functional deficits) can represent intraspecies concordance or
reflect signal strength by demonstrating effects in more than one setting or on more than one
reproductive or developmental process. For example, developmental mortality can be manifested
as early or late resorption, abortion, or stillbirth. If a positive signal in animals is observed at
different stages of development, there is generally greater concern for adverse human

Contains Nonbinding Recommendations
11
reproductive outcomes than if a positive signal is observed only during a single, discrete period.
If the positive signal occurs only for processes that are of limited relevance to humans, there
would be less concern for adverse human outcomes. In addition to its relevance to this evaluation
process, it is also important to define the timing of the period of susceptibility for the observed
positive signal to provide a context for the human risk.
• 	Maternal/paternal toxicity
In weighing a signal of toxicity, the relationship of adverse reproductive or developmental
effects to maternal (and, for fertility studies, paternal) toxicity should be considered when
assessing the level of concern. This assessment is relevant to all endpoints of reproductive and
developmental toxicity. A positive signal occurring at doses that are not maternally toxic
increases concern for human reproductive or developmental toxicity. If a positive signal is
observed only in the presence of frank maternal toxicity, there may be decreased concern,
provided that the positive signal can reasonably be attributed to maternal toxicity (i.e., a causal
relationship between parental toxicity and the signal is established or biologically plausible,3 and
the observed parental toxicity is not expected in humans). When evaluating a positive signal in
two or more species, assessment of the implications of maternal or paternal toxicity should be
based on a composite analysis of the data from all adequately studied species. If any species is
considered inappropriate for assessing the implications of maternal or paternal toxicity, the
evaluation should be performed using the remaining available data.
• 	Dose-response relationship
Concern for human reproductive or developmental toxicity increases when a positive signal is
characterized by any of the following: (1) increased severity of adverse effects with an increase
in dose, (2) increased incidence of adverse effects with an increase in dose, or (3) a high
incidence of adverse effects across all dosed groups. Conversely, there is generally less (or
decreased) concern when these indices of dose-response are absent.
• 	Rare events
Developmental toxicity studies usually lack the statistical power to detect subtle increases in rare
events. Thus, an increased frequency of rare events in drug-exposed animals increases concern
for reproductive or developmental toxicity in humans. The absence of an increased frequency of
rare events, however, does not decrease concern.
• 	Pharmacodynamics: similarity between pharmacologic and toxicologic mechanisms
For drugs with a positive signal, the similarity between the pharmacologic and reproductive or
developmental toxicologic modes of action should be assessed, to the extent that these are
understood. If a positive signal is an extension or progression of, or related response to the drug's
intended pharmacologic mode of action (e.g., delay of parturition by drugs known to suppress
3 The attribution of adverse effects to maternal (or paternal) toxicity can be based on previously collected data
demonstrating the relationship between the maternal/paternal and reproductive or developmental effects.

Contains Nonbinding Recommendations
12
uterine smooth muscle contractility or hypotension in the offspring of dams treated during late
gestation with a drug known to lower blood pressure), there is increased concern for
reproductive or developmental toxicity in humans. There generally is less (or decreased) concern
if the positive signal is attributed to an animal-specific pharmacological response, even though it
may be an extension of the drug's pharmacologic effect (e.g., pregnancy loss in rats due to hypo-
prolactinemia).
• 	Concordance between test species and humans: metabolic and general toxicity profiles
The concordance between the metabolic and drug distribution profiles and general toxicity
profiles in the test species and humans should be evaluated for drugs with a positive signal.
– 	Metabolic and drug distribution profiles
Drug distribution, elimination, and biotransformation (pathways and metabolites) in the
test species and in humans should be compared. Quantitative differences in
metabolic/drug distribution profiles between the test species and humans are often seen,
may not have important implications, and should not be overemphasized. Reproductive
and developmental toxicities in animals induced by compounds whose metabolic and
distribution profiles are very similar in animals and humans increases concern for
reproductive or developmental toxicity in humans.
For compounds with highly dissimilar metabolic or tissue distribution profiles in animals
and humans, there is generally less concern if the toxic effect seen in the test species can
reasonably be attributed to a metabolite or tissue distribution profile not seen in humans.
However, when there are significant differences in drug distribution or metabolic profiles
between several species, yet each test species demonstrates a positive signal for a
reproductive or developmental toxicity, the toxicity is assumed to be attributable to the
parent drug or a common bio-transformed product and concern is increased.
– 	General toxicity profiles
If a drug's overall toxicity profile in one or more test species with a positive signal is
similar to that in humans, there is increased concern for reproductive or developmental
toxicity in humans. If the overall toxicity profiles are dissimilar, there is generally less
(or decreased) concern. When general toxicology data are available for more than one
species, the determination of the level of concern should be based on an assessment of
each test species’ ability to predict human adverse effects in response to the drug.
• 	Relative exposures
When considering the relative exposure comparisons discussed below, more emphasis should be
placed on a parameter within this factor when there is a scientifically plausible link between the
exposure metric and the adverse reproductive or developmental effect. For example, when a
correlation between peak plasma concentration and a developmental effect has been

Contains Nonbinding Recommendations
13
demonstrated for a particular drug, Cmax would be considered the most relevant exposure metric
for making animal to human comparisons for that endpoint.
– 	Kinetic comparison of relative exposure
Comparison of systemic drug exposure at the No Observed Effect Level (NOEL) for the
reproductive or developmental toxicity in the test species to that in humans at the
maximum recommended dose is a critical determination. This comparison should be
based on the most relevant metric (e.g., AUC, Cmax, Cmin, body surface area-adjusted
dose). In general, there is increased concern for reproductive or developmental toxicity in
humans for relative exposure ratios (animal: human) that are <10 and decreased concern
for exposure ratios >25. When applicable, the relative exposure ratio should consider
both the parent compound and its metabolites. For example, it is appropriate to combine
parent and metabolite when both are pharmacologically active and the activity is known
to relate to the reproductive or developmental toxicity. When there are exposure data for
multiple test species, the NOEL exposure for each should be compared to human
exposure at the maximum recommended dose. If the exposure ratios are low (<10 fold) in
multiple species with a positive signal, there is increased concern. If the exposure ratios
are high (>25 fold), there is generally less (or decreased) concern. In the event that a
significant difference in relative exposures is observed between multiple test species, the
appropriateness of the metric (for example, AUC, Cmax) being used to define the
interspecies exposure comparisons should be reassessed. If an alternative metric fails to
reduce the disparity between species, the assessment of concern should be based on the
lowest ratio (i.e., in the most sensitive species).
Relative interspecies exposure data should be evaluated in light of possible species
differences in protein binding (free drug concentration), receptor affinity (if related to the
positive signal) or site-specific drug concentrations. In the absence of meaningful
differences between the test species and humans in these parameters, the interspecies
comparisons should be based on total drug exposure.
– 	Biomarkers as a measure of relative exposure
The purpose of this relative exposure metric is to compare the dose causing reproductive
or developmental toxicity in the test species to the therapeutic dose in humans,
normalized to the doses causing a response common to both species. In practice, this is
done by taking the exposure at the NOEL for the adverse reproductive or developmental
effect and dividing by the exposure at which the biomarker response is seen in the test
species. This is compared to the human therapeutic exposure divided by the exposure at
which the biomarker response is seen in the human. The ratio calculated for animals is
then divided by the ratio calculated for humans. When this ratio of relative biomarker
exposure (animal: human) is < 10, there is generally increased concern for human
reproductive or developmental toxicity. When this ratio is > 25, there is generally less
concern.

Contains Nonbinding Recommendations
14
When there are data to compute relative biomarker exposure ratios for multiple species,
the level of concern assessment should be based on an integrated analysis of data from all
adequately studied species. When there are non-concordant biomarker ratios between
multiple test species, the relevance of the biomarker as expressed in the various species
should be considered before making an assessment. If there is no scientific rationale for
the disparity between species, the biomarker, as a measure of exposure, will be of
questionable utility.
• 	Class Alerts
Consideration of a class associated effect should be based on prior human experience for a drug
with closely related chemical structure (parent or metabolite) or related pharmacologic effect and
with known reproductive or developmental outcomes in humans. There is increased concern for
reproductive or developmental toxicity in humans when the drug is from a class of compounds
known to produce adverse effects in humans and animals. There is decreased concern only in
circumstances in which a class of compounds, although demonstrating adverse effects in
animals, has previously been shown definitively to have no related adverse effects on human
reproduction or development.
3. 	Summary/Integration of Positive Findings
When there is a positive finding in nonclinical studies for one or more endpoints of reproductive
or developmental toxicity, there is a potential for increased human risk. Multiple considerations
contribute to the overall evaluation of the nonclinical data and conclusions regarding human risk.
These include factors that can modify the level of concern for human adverse effects determined
from the nonclinical signal. Factors can increase or decrease concern, and some factors can carry
greater weight than others. Positive signals should be evaluated to estimate the likelihood of
increased reproductive or developmental risk for humans using the following general procedure:
• 	In evaluating the level of increased risk, all relevant information should be considered,
including nonclinical reproductive and general toxicology data and human and animal
pharmacodynamic and pharmacokinetic data.
• 	Factors that may affect the level of concern associated with a positive signal of
reproductive or developmental toxicity should be assessed.
• 	The analysis should take into account the quality and type of data under consideration.
• 	A weight of evidence approach should then be applied to arrive at an overall conclusion
for reproductive or developmental toxicity (Figure C). The following are examples of
possible summary risk conclusions for the evaluation:
– 	Does Not Appear to Increase Risk: The drug is not anticipated to increase the risk
of adverse developmental (or reproductive) outcomes in humans when used in
accordance with dosing information in the product label.

Contains Nonbinding Recommendations
15
– 	May Increase Risk: The drug may increase the risk of adverse developmental (or
reproductive) outcomes in humans when used in accordance with the dosing
information in the product label.
– 	Predicted to Increase Risk: The drug is expected to increase the risk of adverse
developmental (or reproductive) outcomes in humans when used in accordance
with the dosing information in the product label.
The factors discussed above are derived from a limited sample of pharmaceuticals for which the
clinical outcomes are reasonably well defined and are not considered exhaustive. CDER believes
that using the various factors to assess a drug’s potential to increase the risk of adverse
reproductive and developmental outcomes in humans will result in a more accurate as well as
more unbiased and uniform evaluation. CDER also believes this approach will help identify
specific areas in which additional information about a pharmaceutical would be useful in more
fully defining risk and would enable specific analyses of areas of disagreement that influence the
risk evaluation.

Contains Nonbinding Recommendations
16
GLOSSARY
ADME − Αbsorption, distribution, metabolism, and elimination
Class Alert – A structurally adverse reproductive or developmental effect previously
demonstrated in humans by related chemical entities or compounds with similar
pharmacodynamic effects
Developmental Toxicity − Αny adverse effect induced prior to attainment of adult life. It
includes effects induced or manifested in the embryonic or fetal period and those induced or
manifested postnatally. These are divided into four major manifestations or classes: mortality,
dysmorphogenesis, alterations to growth, and functional toxicities.
Factor – For purposes of this guidance, a factor is one of the various considerations used to
evaluate the level of concern for a positive signal of developmental or reproductive toxicity. The
factors include (1) Cross-species concordance of reproductive or developmental effects; (2)
Multiplicity of effects; (3) Maternal/paternal toxicity; (4) Dose–response relationship; (5) Rare
events; (6) Pharmacodynamics: similarity between pharmacologic and toxicologic mechanisms;
(7) Concordance between test species and humans: metabolic and general toxicity profiles; (8)
Relative exposure; and (9) Class alerts. They are summarized in Figure C and are discussed in
Section C.
Fertility – Reproductive competence
Lactation − Τhe secretion of milk or the period of milk secretion
Malformation − Α permanent alteration (abnormality) in which there is a morphologic defect of
an organ or a larger region of the body, resulting from an abnormal developmental process. They
generally occur at a low frequency in the control population and/or will adversely affect survival,
growth, or development of functional competence
Maternal (Paternal) Toxicity – Toxicity to the mother (maternal) or the father (paternal) in a
reproductive toxicology study, but not necessarily a toxicity to reproductive function
Parturition – Labor and delivery
Positive Signal − Α treatment-related reproductive or developmental toxicity
Rare Event – An endpoint that occurs in less than 1 percent of the control animals in a study
and in historical control animals
Reproductive Toxicity − Structural and/or functional alterations that may affect reproductive
competence in sexually mature males and females. These may be manifested as impairment of
fertility, parturition, or lactation.

Contains Nonbinding Recommendations
17
Structural abnormalities – Includes malformations and variations
Toxicologic Effect − Αny adverse effect of a therapeutic agent
Variation − An alteration that may occur at a relatively high frequency and/or represents a
retardation in development, a transitory alteration, or a permanent alteration not believed to
adversely affect survival, growth, development, or functional competence

Contains Nonbinding Recommendations
18
FIGURES

Contains Nonbinding Recommendations
19

Contains Nonbinding Recommendations
20