EMA发布人用药品人类生殖与哺乳风险评估指南修订1草案并启动公众咨询
Draft guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product information - Revision 1
EMA的CHMP与PRAC于2026年11月3日通过并发布人用药品人类生殖与哺乳风险评估指南修订1草案(EMA/CHMP/PRAC/203927/2005 Rev.1),公众咨询自2026年11月9日起至2027年2月28日截止。
作为2005年版指南的修订草案,文件给出从数据到产品信息的风险评估流程并开放征求意见,便于对照现行版本定位变化。
PDF 文字版;图形和原始排版请参阅官方 PDF。
第 1 页
1 03 November 2026
2 EMA/CHMP/PRAC/203927/2005 Rev.1
3 Committee for Medicinal Products for Human Use (CHMP)
4 Pharmacovigilance Risk Assessment Committee (PRAC)
5 Guideline on risk assessment of medicinal products in
6 human reproduction and breastfeeding: from data to
7 product information
8 Draft
9
Draft agreed by Committee for Medicinal Products for Human Use 13 April 2026
(CHMP); Pharmacovigilance Risk Assessment Committee (PRAC)
Adopted by CHMP and PRAC for release for consultation 03 November 2026
Start of public consultation 09 November 2026
End of consultation (deadline for comments) 28 February 2027
10
11 This guideline replaces “Guideline on Risk Assessment of Medicinal Products on Human Reproduction
12 and Lactation: from Data to Labelling (EMEA/CHMP/203927/2005)”
13
Comments should be provided using this EUSurvey form. For any technical issues, please contact
the EUSurvey Support.
14
Keywords Pregnancy, lactation, breastfeeding, fertility, contraindication, risk
minimisation, non-clinical assessment, clinical assessment, risk
assessment, Summary of Product Characteristics, Package Leaflet.
15
16
Official address Domenico Scarlattilaan 6 ● 1083 HS Amsterdam ● The Netherlands
Address for visits and deliveries Refer to www.ema.europa.eu/how-to-find-us
Send us a question Go to www.ema.europa.eu/contact Telephone +31 (0)88 781 6000 An agency of the European Union
© European Medicines Agency, 2026. Reproduction is authorised provided the source is acknowledged.
第 2 页
17 Guideline on risk assessment of medicinal products in
18 human reproduction and breastfeeding: from data to
19 product information
20 Table of contents
21 Executive summary ................................................................................................. 4
22 1. Introduction (background) .............................................................................. 5
23 2. Scope ............................................................................................................... 5
24 3. Legal basis and relevant guidelines .................................................................. 6
25 4. Non-clinical assessment of risk to human reproduction and development ....... 6
26 4.1. Introduction......................................................................................................... 6
27 4.1.1. Choice of species ............................................................................................... 7
28 4.1.2. New approach methods ...................................................................................... 7
29 4.1.3. Route of administration ...................................................................................... 7
30 4.1.4. Dose regimen/scheme ........................................................................................ 7
31 4.1.5. Pharmacokinetics/toxicokinetics .......................................................................... 8
32 4.1.6. Mechanism ....................................................................................................... 8
33 4.1.7. Multigenerational studies .................................................................................... 9
34 4.2. Evaluation process................................................................................................ 9
35 4.2.1. Uncertainties from available data ......................................................................... 9
36 4.2.2. No effect detected ............................................................................................. 9
37 4.2.3. Effects detected or known ................................................................................... 9
38 5. Clinical assessment of risk to human reproduction and development ............ 11
39 5.1. Introduction....................................................................................................... 11
40 5.2. Effects on fertility ............................................................................................... 11
41 5.3. Effects on pregnancy .......................................................................................... 12
42 5.3.1. Adverse pregnancy outcomes ............................................................................ 12
43 5.3.2. Clinical assessment of data on pregnancy outcomes............................................. 14
44 6. Assessment regarding breastfeeding ............................................................. 17
45 6.1. Introduction....................................................................................................... 17
46 6.2. Risk assessment for a breastfed infant .................................................................. 17
47 6.2.1. Major elements of assessment: ......................................................................... 17
48 6.2.2. Complementary data for assessment: ................................................................ 18
49 6.3. Assessment of effect of a medicinal product on lactation (breastmilk production, secretion
50 and composition) ...................................................................................................... 18
51 6.4. Proteins including monoclonal antibodies and risk assessment for breastfeeding ........ 18
52 7. Integration: From data to product information .............................................. 19
53 7.1. General principles ............................................................................................... 19
54 7.1.1. Fertility .......................................................................................................... 20
55 7.1.2. Pregnancy ...................................................................................................... 20
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 2/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 3 页
56 7.1.3. Breastfeeding .................................................................................................. 22
57 7.2. Principles for recommendations ............................................................................ 23
58 7.2.1. Fertility .......................................................................................................... 23
59 7.2.2. Pregnancy ...................................................................................................... 23
60 7.2.3. Individuals of reproductive potential .................................................................. 25
61 7.2.4. Breastfeeding .................................................................................................. 26
62 8. Reflections for the product information.......................................................... 27
63 8.1. Introduction....................................................................................................... 27
64 8.2. Summary of product characteristics (SmPC) .......................................................... 27
65 8.2.1. Section 4.6 of the SmPC ................................................................................... 28
66 8.2.2. Section 4.3 of the SmPC ................................................................................... 30
67 8.2.3. Section 4.4 of the SmPC ................................................................................... 30
68 8.2.4. Section 4.5 of the SmPC ................................................................................... 30
69 8.3. Package leaflet (PL) ............................................................................................ 30
70 9. References ..................................................................................................... 32
71 Annex ................................................................................................................... 34
72
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 3/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 4 页
73 Executive summary
74 Integrating non-clinical and human data to assess risks of medicinal products for human reproduction,
75 development and if used during breastfeeding is essential. Reflecting this assessment and subsequent
76 recommendations, in the product information of a medicinal product is also important to support
77 informed decisions by health care professionals and individuals using the medicinal product, and to
78 reduce adverse outcomes following use in these populations.
79 This is the first revision of the Guideline on Risk Assessment of Medicinal Products on Human
80 Reproduction and Lactation: from Data to Labelling. The main aim of this revision was to reflect
81 current standards regarding non-clinical - including the option for new approach methods (NAMs) - and
82 clinical assessments. The latter relate to different types of clinical studies, including
83 pharmacoepidemiologic studies, as well as other types of post marketing data. Further details on risk
84 classifications and how these impact recommendations for use of medicinal products during pregnancy,
85 in individuals of reproductive potential, and in breastfeeding are included. Examples of standard texts
86 (statements) for the Summary of Product Characteristics (SmPC) have been updated, and new
87 standard statements on fertility, as well as new standard texts for the Package Leaflet (PL) have been
88 added.
89 The guideline describes a stepwise process, starting with two sections on assessment of non-clinical
90 (section 4) and clinical data regarding human reproduction and development (section 5), followed by a
91 section on assessment of non-clinical and clinical data in relation to breastfeeding (section 6).
92 Thereafter, a section on integration of non-clinical and clinical data follows (section 7.1), which
93 describes general principles including classification of risks across areas, and provides guidance on
94 recommendations for use for each specific area (section 7.2). The guideline ends with a section on
95 reflections for the product information (section 8, both the SmPC, and PL) as well as an Annex
96 covering standardised texts for data description, recommendations and risk minimisation measures
97 in relation to pregnancy, individuals of reproductive potential and breastfeeding. Figure 1 describes the
98 structure of the guideline and the process it is intended to cover.
99 Figure 1. Stepwise process of risk assessment: from data to product information
100
101
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 4/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 5 页
102 1. Introduction (background)
103 The information and recommendations on how to use a medicinal product in relation to human
104 reproduction, pregnancy or breastfeeding should enable health care professionals to manage different
105 circumstances, and support individuals in making informed decisions. These situations include
106 prescribing medicinal products to individuals of reproductive potential 1, or pregnant and breastfeeding
107 individuals. Therefore, the product information (PI), specifically the SmPC and PL, should present the
108 available evidence on identified and potential risks related to fertility, pregnancy, and breastfeeding,
109 together with recommendations on the use of the medicinal product in these populations as
110 appropriate, and on the management of risks in clinical practice.
111 Evaluation of effects on human fertility (male and female) frequently relies on findings from non-
112 clinical studies (reproductive toxicity studies and/or repeated dose toxicity studies), along with
113 knowledge on the mechanism of action and relevant clinical evidence of the active substance(s).
114 For pregnancy, the risk assessment is based on an integrated and continuous evaluation of clinical
115 experience and non-clinical toxicology, pharmacology and pharmacokinetic data.
116 During medicinal product development, and early post marketing period, non-clinical data are of great
117 importance since clinical experience is still lacking or limited in most clinical settings. In certain
118 situations, pregnant and/or breastfeeding individuals may be included in clinical trials [1], depending
119 on the medical condition being studied, as well as on the potential risk profile of the substance, or such
120 studies are undertaken with an aim to characterise use of a medicinal product in pregnant individuals.
121 Generally, clinical data becomes more abundant in the post marketing phase as real-world experience
122 with the medicinal product increases, and such data are essential to fully characterise safety in
123 humans. When formulating recommendations with respect to use in females of reproductive potential
124 and during pregnancy, the potential risk of adverse embryo-fetal or infant outcomes due to in utero
125 exposure should be weighed against the risk of not treating the pregnant individual.
126 For breastfeeding, the potential risk to the infant should be weighed against the known benefits of
127 breastfeeding itself. Risk assessment should preferably be based on human data, including
128 concentrations of the active substance (and/or metabolites) in breast milk, estimated exposure of the
129 breastfed infant, the toxicity profile of the substance and suspected adverse effects in breastfed
130 infants. Data on pharmacokinetic properties and certain non-clinical data can support the assessment.
131 2. Scope
132 This guideline applies to medicinal products for human use. The purpose of this guideline is to:
133 • Describe the process for integrating non-clinical and clinical data, and to highlight key factors in
134 assessing the risk of adverse reproductive and developmental effects in humans. This includes
135 effects on fertility and on pregnancy, including effects on the embryo/fetus or child health following
136 exposure during pregnancy, and effects on the breastfed infant.
137 • Outline how to describe available data, including potential or identified risks, and recommendations
138 for the use of a medicinal product, including relevant risk minimisation measures, in the SmPC and
139 the PL. Standard statements are provided.
1 In this guideline, terminology relating to sex and gender is selected according to the context and population described.
Sex-specific terms (e.g. female, male) are used where necessary for the understanding, such as fertility or for individuals of
reproductive potential. Gender-neutral terms (e.g. individuals) are used when referring to those who are pregnant or
breastfeeding. “Females of reproductive potential” refers to all females who are able to become pregnant, including
adolescents. The term “individual” refers to a person using, or considering the use of, a medicinal product for treatment or
prevention.
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 5/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 6 页
140 3. Legal basis and relevant guidelines
141 This Guideline should be read in conjunction with the introduction and general principles of Annex I to
142 Directive 2001/83/EC, as amended, and all other relevant EU and ICH guidelines. These include, but
143 are not limited to:
144 • Preclinical safety evaluation of biotechnology-derived pharmaceuticals
145 (EMA/CHMP/ICH/731268/1998 (ICH S6(R1))
146 • Guideline on detection of reproductive and developmental toxicity for human pharmaceuticals
147 (EMA/CHMP/ICH/544278/1998 (ICH S5)
148 • Guideline on genotoxicity testing and data interpretation for pharmaceuticals intended for human
149 use (EMA/CHMP/ICH/126642/2008 (ICH S2)
150 • Toxicokinetics: A Guidance for Assessing Systemic Exposure in Toxicology Studies
151 (CPMP/ICH/384/95 (ICH S3A)).
152 • Guideline on non-clinical testing for inadvertent germline transmission of gene transfer vectors
153 (EMEA/273974/2005)
154 • SWP/NcWP recommendations on the duration of contraception following the end of treatment with
155 a genotoxic drug (EMA/CHMP/SWP/74077/2020)
156 • ICH guideline M3(R2) on non-clinical safety studies for the conduct of human clinical trials and
157 marketing authorisation for pharmaceuticals (EMA/CPMP/ICH/286/1995)
158 • ICH guideline M3 (R2) - questions and answers (EMA/CHMP/ICH/507008/2011)
159 • Guideline on inclusion of pregnant and breastfeeding individuals in clinical trials
160 (EMA/CHMP/ICH/149462/2025 (ICH E21)
161 • Guideline on good pharmacovigilance practices (GVP): Product- or population-specific
162 considerations III: Pregnant and breastfeeding women and their children exposed in utero or via
163 breastmilk (EMA/653036/2019)
164 • Recommendations related to contraception and pregnancy testing in clinical trials, CTCG Clinical
165 Trials Coordination Group
166 • GVP Module V – Risk management systems (EMA/838713/2011)
167 • GVP Module XVI Addendum I – Risk minimisation measures for medicinal products with embryo-
168 fetal risks (EMA/608947/2021)
169 • ENCePP Guide on Methodological Standards in Pharmacoepidemiology (EMA/95098/2010)
170 • ENCePP Annex 2 to the Guide on Methodological Standards in Pharmacoepidemiology
171 • Guideline on the Summary of Product Characteristics. Rules governing Medicinal products in the
172 European Community Volume 2C The Notice to Applicants
173 4. Non-clinical assessment of risk to human reproduction
174 and development
175 4.1. Introduction
176 Developmental and reproductive toxicity (DART) studies aim to detect effects of a medicinal product on
177 mammalian reproduction and development. They typically consist of a fertility and early embryonic
178 development (FEED) study, embryo-fetal development (EFD) studies in two species (rodent and non-
179 rodent), and a pre- and postnatal development (PPND) study. However, this standard approach may
180 be adapted according to various factors including the type of medicinal product, target population,
181 knowledge of pharmacological target(s) and its(their) role in reproduction and development.
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 6/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 7 页
182 Results of other non-clinical studies and weight of evidence (WoE) approaches should also be
183 considered as they may significantly influence the testing strategy and/or risk assessment. See ICH
184 S5(R3) guideline [2] for further details on these aspects. Regarding risk assessment for breastfeeding,
185 see section 6.
186 For the non-clinical assessment, reproductive and developmental toxicity includes:
187 • Effects on reproduction in the parental generation (i.e. fertility): any adverse effects (structural
188 and functional) on the male and/or female reproductive organs or on the related endocrine
189 systems including amongst others: gamete production and transport, reproductive cycle, mating
190 behaviour, or modification of other functions that are dependent on the integrity of the
191 reproductive system.
192 • Effects on development in the progeny: any adverse effects (structural and functional) on the
193 conceptus resulting from prenatal exposure or induced during the suckling period or manifested
194 postnatally (adverse effects on e.g. behaviour and / or sexual maturation).
195 4.1.1. Choice of species
196 The rat is a common rodent species in all types of DART studies (FEED, EFD and PPND), whereas the
197 rabbit is a common non-rodent second species for the EFD study. For the EFD studies, at least one of
198 the selected species should be responsive to the primary pharmacodynamic effect of the medicinal
199 product. In addition, the pharmacokinetic profile, including biotransformation, should enable
200 assessment of the relevance of the species selected for DART testing and thus an extrapolation of the
201 results from animals to humans.
202 In cases where the rat or rabbit are considered unsuitable for testing due to e.g. lack of
203 pharmacological and/or pharmacokinetic relevance or tolerability issues, an alternative rodent or non-
204 rodent species should be considered. Other options including the use of genetically modified animals or
205 a species-specific surrogate molecule may also be considered provided that pharmacological relevance
206 is ensured. When no relevant species can be identified, the impact of this deficiency should be
207 discussed in the assessment report and considered for the recommendations in the SmPC and PL.
208 4.1.2. New approach methods
209 NAMs, including alternative in vitro, ex vivo, and non-mammalian in vivo assays may be valuable to
210 identify adverse effects or to contribute to the understanding of the mechanisms underlying
211 developmental and reproductive toxicity at any stage of medicinal product development. These
212 methods when qualified for context of use may also support risk assessment and, where justified,
213 reduce or replace certain in vivo studies. Further details are provided in the ICH S5(R3) guideline [2].
214 4.1.3. Route of administration
215 The route of administration used in DART studies should generally be the one intended for human use.
216 An alternative route may be acceptable if a similar or ‘better’ pharmacokinetic profile (i.e. resulting in
217 higher exposure) can be demonstrated.
218 4.1.4. Dose regimen/scheme
219 The dose range tested in DART studies should have covered a dose resulting in minimal maternal
220 toxicity such as e.g. decreased body weight or food consumption or, at the high dose, result in a 25-
221 fold exposure margin compared with the estimated human exposure at the maximum recommended
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 7/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 8 页
222 dose. See ICH S5 guideline [2] for more details. While determining a “no observed adverse effect
223 level” (NOAEL) is desirable, dose levels should be chosen to reveal dose-related trends. Dose levels
224 leading to serious maternal and/ or offspring toxicity, including mortality, can mask a potential for
225 developmental toxicity.
226 It should be evaluated whether the applied dose intervals/frequencies achieve sufficient exposure
227 during the sensitive periods under study and are clinically relevant.
228 4.1.5. Pharmacokinetics/toxicokinetics
229 Exposure in pregnant animals to the active substance(s) and/or major human metabolite(s) (≥10%)
230 [2-5], should be assessed. Indirect estimation of exposure in pregnant animals, based on toxicokinetic
231 data from non-pregnant animals from the same species used in different repeated toxicity studies
232 represents an element of uncertainty.
233 When available, information regarding placental transfer of the active substance (and/or metabolites)
234 can be of value for the assessment.
235 4.1.6. Mechanism
236 Toxicants can disrupt reproduction and/or development through maternal/paternal toxicity, direct
237 effects on the conceptus (placenta, extraembryonic membranes, or embryo/fetus), indirectly via
238 maternally mediated effects, or a combination of both. Thus, substances not able to cross the placenta
239 can still cause embryo-fetal toxicity due to indirect alterations of maternal physiology (e.g.
240 hyperglycaemia).
241 If reproductive and/or developmental toxicity has been identified, an understanding of the molecular
242 basis (i.e. mechanism of action) to support a biologically plausible explanation for the observed
243 toxicity, together with interspecies comparisons of the pharmacological effect, is of value for risk
244 assessment. Mechanistic data can include information on pharmacokinetics, interactions with fertility
245 and developmental processes in the mother (hormonal disruption, cytotoxic effects, inhibition of
246 angiogenesis essential for ovarian function, placental development, embryonic growth), embryo/fetus
247 or extraembryonic tissues. Since medicinal products are often targeted towards a well-defined
248 biological entity (receptor, ion-channel, protein etc.), the on-target molecular interaction is usually
249 known. However, the actual embryo-fetal exposure, including time point for exposure in relation to the
250 sensitive period of development of a specific organ/tissue, will influence the final outcome.
251 A WoE evaluation (see section 4.2.) is an important part of human risk assessment since it is
252 acknowledged that it can be difficult to generate data in sufficient detail for a clear understanding of
253 both mode of action, and mechanism of action behind embryo-fetal toxicity, especially for a new
254 chemical entity from a new class, and within the remit of regulatory testing. This is due to the
255 complexity of normal development, a multitude of interaction points for a substance, and since mode
256 of action understanding involves several levels of biological organisation.
257 For gene therapies or vaccines using genetically modified viral vectors that are replication-
258 incompetent, the risk of vertical maternal and paternal germline transmission of vector DNA, should be
259 assessed. See Guideline on non-clinical testing for inadvertent germline transmission of gene transfer
260 vectors [6] for more details.
261 4.1.6.1. Genotoxicity
262 Results of comparative studies have shown that, in a qualitative sense, most germ cell mutagens are
263 likely to be detected as genotoxic in somatic cell tests. If positive genotoxic effects are identified in the
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 8/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 9 页
264 genotoxicity battery tests (See ICH S2 [7]), this also raises concerns for reproductive and
265 developmental effects (both at gamete level as well as the embryo, depending on the exposure time).
266 See “Safety Working Party (SWP) recommendations on the duration of contraception following the end
267 of treatment with a genotoxic drug (EMA/CHMP/SWP/74077/2020)” [8] for further details.
268 4.1.7. Multigenerational studies
269 Reproductive studies evaluating drug-related effects over generations may be available in exceptional
270 circumstances. Particular attention should be paid to the robustness of the study design to meet the
271 study objectives driven by substance specific potential safety concerns. The clinical relevance of the
272 reported findings should be considered in the context of all other available non-clinical and clinical data
273 (including mechanistic studies).
274 4.2. Evaluation process
275 The non-clinical assessment should be based on all pharmacological (primary, secondary and safety),
276 pharmacokinetic and toxicological data available. In some situations, all elements from standard
277 reproductive toxicity testing may not be available or are not warranted. A WoE approach can in these
278 cases be adequate to support decision-making when evaluating reproductive and developmental risk.
279 Such approaches use the totality of evidence available, and take into account literature data, class
280 effect data and data from alternative models or NAMs such as (complex) in vitro models or in silico
281 models, as well as all available relevant non-clinical and clinical data for the substance. The severity of
282 the disease, therapeutic indication, the target population and other relevant risk factors should also be
283 taken into account.
284 In this context, WoE approaches are particularly useful to demonstrate presence of risk. However, if
285 the data are sufficiently robust and substantiated, a WoE evaluation to support the likelihood of
286 absence of risk can be accepted on a case-by-case basis.
287 4.2.1. Uncertainties from available data
288 The representativeness, completeness, strengths and weaknesses of available data sets should always
289 be considered in the evaluation process. If some non-clinical studies (including reproductive toxicity)
290 are lacking, or some studies are irrelevant, or if data on certain endpoints are lacking, the impact of
291 such insufficient data should be addressed in the evaluation and considered for the overall conclusion
292 on developmental and reproductive risk.
293 4.2.2. No effect detected
294 If appropriately designed developmental and reproductive toxicity studies with adequate systemic
295 exposure do not indicate direct or indirect harmful effects, and there are no relevant effects identified
296 in other pharmacological or toxicological studies, this conclusion enables finalisation of the non-clinical
297 evaluation process.
298 If no findings are observed for a substance belonging to a class where adverse effects are expected,
299 this may need further clarification.
300 4.2.3. Effects detected or known
301 If an effect suggesting reproductive and/or developmental toxicity is detected, the evaluation process
302 should continue as outlined below in order to assess the level of concern.
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 9/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 10 页
303 There are several contributory elements that may affect the level of concern. The following aspects
304 should be taken into consideration:
305 4.2.3.1. General aspects
306 • Recognition of an effect
307 Statistically significant differences between treatment and control groups alone do not necessarily
308 indicate a safety concern, nor does lack of statistical significance impute absence of an effect. Other
309 criteria of importance include biological plausibility, reproducibility, medicinal product- or species-
310 specific mode of action, relationship to an animal-specific metabolite, the magnitude and nature of the
311 effect and/or clear dose-response relationship. Furthermore, reproductive toxicity studies usually lack
312 statistical power to detect subtle increases in rare events. Thus, an increased frequency of rare events
313 in exposed animals in comparison with the study control as well as with recent historical control data
314 from the same strain and preferably from the laboratory in which the study was performed, may
315 increase the concern for reproductive toxicity in humans.
316 • Inter-species concordance
317 Even if adverse findings from only one species can raise serious concerns, if similar effects are
318 observed in more than one species, it may increase the concern for developmental and/or reproductive
319 toxicity in humans. Inter-species concordance is most likely to be identified for structural birth
320 defects/congenital anomalies, or histopathological examinations of reproductive organs, because these
321 endpoints are studied in multiple species.
322 • Type of effects
323 Structural birth defects/malformations and embryo-fetal lethality have more weight in estimating a
324 concern for developmental toxicity in humans, than general effects (e.g. growth retardation and
325 variations, as delays in ossification). The latter, possibly reversible effect(s), might be induced by
326 maternal stress or other maternal toxicity.
327 Data from reproductive toxicity studies and/or from repeated dose toxicity studies in sexually mature
328 animals may suggest detrimental effects on fertility. In repeated dose toxicity studies, such effects
329 could be alterations to endocrine function or histopathological changes in the gonads or genital organs,
330 which may be detected in both rodents and non-rodents. Further understanding of possible
331 mechanisms is desirable for risk assessment.
332 • Multiplicity of effects
333 Multiplicity of effects usually increases concern for reproductive toxicity in humans. Multiplicity refers to
334 the observation, in a single species or animal model, of two or more effects with related biological
335 endpoints in fetal examinations, e.g. abnormally shaped vertebrae plus abnormal number of vertebrae
336 would give rise to increased concern, more than just a single observation with regards to vertebral
337 development.
338 • Adverse effects at various stages of the reproductive process
339 This refers to observations in a single species or animal model, of effects during different stages of the
340 reproductive process, e.g. increased early intrauterine death plus some malformations would give rise
341 to increased concern with regards to developmental toxicity.
342 • Maternal toxicity
343 If developmental toxicity is seen only at maternally toxic dose levels, further understanding of
344 potential causality between maternal and developmental toxicity is desirable. From a regulatory
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 10/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 11 页
345 perspective, claims that developmental toxicity is due to maternal toxicity should be adequately
346 justified. Further, there needs to be support that such findings do not have clinical relevance (e.g., by
347 providing mechanistic data in support of maternal toxicity being the reason for the observed
348 developmental toxicity).
349 4.2.3.2. Specific aspects
350 • Animal to human exposure ratio
351 This should be based on the systemic exposure of the animal at the NOAEL for reproductive toxicity
352 and the systemic exposure in human at the maximum therapeutic dose. In case of serious embryo-
353 fetal outcomes in the animal studies, estimations of animal and human exposure ratios at doses
354 associated with malformation or embryo-fetal lethality can provide further insight for the risk
355 assessment.
356 When reporting animal data in SmPC section 4.6, a tiered system on how to interpret NOAEL for EFD
357 toxicity testing should be taken into account [2]. In general, there is increased concern when the
358 NOAEL occurs at exposures less than 10-fold the human exposure at the maximum recommended
359 dose. Above this threshold, concern is reduced, and effects observed at 25-fold or larger exposure
360 margins are usually of minor concern for clinical use.
361 • Neurodevelopmental risk assessment
362 The entire nervous system, including the central nervous system (CNS) undergoes continuous complex
363 multifactorial development from early embryogenesis through to adulthood. From a non-clinical
364 perspective, the risk assessment should take into account the totality of available evidence namely
365 pharmacology, pharmacokinetics and toxicology data. Nevertheless, non-clinical models may not fully
366 reflect the complexity of human neurodevelopment, which needs to be considered in the assessment.
367 In particular, the clinical relevance of finding(s) from neurobehavioral tests conducted in the PPND
368 studies, should be thoroughly interpreted in combination with the overall human risk assessment.
369 5. Clinical assessment of risk to human reproduction and
370 development
371 5.1. Introduction
372 The clinical assessment will generally be based on various types of clinical data. These could include
373 case reports, case series, clinical trials, and epidemiological studies and originate from different
374 sources, e.g. electronic healthcare records, pregnancy registries, registries of congenital anomalies,
375 and spontaneously reported post-marketing data.
376 The assessment should consider the scientific quality and robustness of available data, methodology
377 for data collection, study designs, the clinical context such as target population, and the amount of
378 available data.
379 5.2. Effects on fertility
380 Fertility refers to the ability of an individual (male or female), a couple, or a population to reproduce.
381 Infertility is defined as a disease of the male or female reproductive system characterized by the failure
382 to achieve a clinical pregnancy after one year or more of regular unprotected sexual intercourse [9].
383 Clinical evaluations of fertility typically focus on couples who meet this criterion.
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 11/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 12 页
384 Medicinal products may impair fertility of males and/ or females, through various mechanisms such as
385 hormonal disruption or cytotoxic effects. Effects may be reversible or permanent. The pharmacological
386 properties of the medicinal product, together with the characteristics of the target population, will
387 guide the most appropriate approach to assessing potential adverse effects on fertility.
388 Evaluation of fertility may include, but is not limited to, assessing effects on hypothalamic/pituitary
389 gonadal function, as well as effects on the reproductive organs/tissues. In addition, impairment of
390 implantation or an increased occurrence of early spontaneous abortion, are other relevant fertility
391 markers. Surrogate markers of fertility may be appropriate and can be studied on a case-by-case
392 basis. Studying effects on fertility by monitoring reproductive outcome in couples is generally
393 considered difficult due to practical and methodological challenges.
394 5.3. Effects on pregnancy
395 5.3.1. Adverse pregnancy outcomes
396 The administration of a medicinal product during pregnancy may cause various adverse effects in the
397 (unborn) child, including morphological, and/or functional developmental disturbance in the embryo or
398 fetus - whether detected at birth or not – as well as embryonic or fetal death. The nature and severity
399 of these effects depend on different factors including timing of exposure during critical phases of
400 development. Also, the maternal disease being treated may in itself affect embryo/fetal development
401 adversely, which needs to be taken into account.
402 Main types of adverse pregnancy outcomes include:
403 • Non-live birth, including spontaneous abortion (miscarriage) and fetal death
404 • Congenital malformation
405 • Embryotoxic and/or fetotoxic effects
406 • Neonatal outcomes
407 • Neurodevelopmental disorders (NDD)
408 In addition, some effects fall outside these categories (e.g. transplacental carcinogenesis). For detailed
409 description/definition of “susceptible periods” and “adverse pregnancy outcomes” see “Guideline on
410 good pharmacovigilance practices (GVP) Product- and Population- Specific Considerations III: Pregnant
411 and breastfeeding women and their children exposed in utero or via breastmilk” [10].
412 5.3.1.1. Non-live birth, including spontaneous abortion (miscarriage) and fetal death
413 Non-live births include spontaneous abortion/miscarriage (early fetal death, that occurs before 20
414 completed weeks of gestation), stillbirth (late fetal death, that occurs after 20 completed weeks of
415 gestation). Also, termination of pregnancy due to embryo-fetal abnormalities is included in this group
416 of events. Incomplete capture of early pregnancies because pregnancies may not be immediately
417 recognised makes precise estimation of spontaneous abortion frequencies uncertain but is about 25%
418 of all pregnancies. This frequency increases with maternal age. Stillbirths are reported to occur in
419 about 0.3% of total births in Europe (range: 1.8-4.7 per 1000) [11].
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 12/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 13 页
420 5.3.1.2. Congenital malformation (birth defects)
421 Congenital malformation 2 are structural anomalies of organs, systems, or parts of the body that occurs
422 during intrauterine development and may be associated with altered organ or system function. A
423 congenital malformation may result from genetic factors or from exposures to an environmental factor
424 including medicinal products. Congenital malformations can have significant medical, surgical, or
425 cosmetic consequences.
426 While it may be possible to diagnose some congenital malformations prenatally, others will only be
427 captured during the first years of life, leading to higher prevalences with longer follow-up after birth.
428 Additionally, in studies restricted to live births, risks may be underestimated if the medicinal product
429 caused pregnancy losses (i.e. live-birth bias, see [12,13]), or if the pregnancy is terminated due to
430 major malformations detected in utero. Prevalence also varies depending on how major malformations
431 are defined in the data source. In Europe, the EUROCAT classification system provides a harmonised
432 framework for categorizing congenital malformations across data sources, assigning major or minor
433 status to specific conditions within organ systems [14]. Additional factors, such as the inclusion or
434 exclusion of minor malformations or those due to genetic diseases, will also impact reported
435 prevalence. Prevalence estimates may also differ depending on whether exposure information is
436 collected prospectively (before pregnancy outcome is known, either adverse or non-adverse) or
437 retrospectively (after the outcome), the latter being more susceptible to recall bias. The live birth
438 prevalence of major congenital malformations is reported to be 2–4% in Europe [14] and 3-5% in the
439 United States [15].
440 5.3.1.3. Embryotoxic and/or fetotoxic effects
441 Embryotoxic and/or fetotoxic effects include impaired fetal growth (low/very low birth weight, small for
442 gestational age or intrauterine growth restriction), excessive growth (large for gestational age or fetal
443 macrosomia), and adverse effects on organs (e.g., endocrine or reproductive organs). Embryotoxicity
444 refers to effects from exposure during the first trimester, while fetotoxicity generally refers to exposure
445 in the second and/or third trimester.
446 5.3.1.4. Adverse neonatal outcomes
447 Adverse neonatal outcomes generally reflect pharmacological effects of the medicinal product during
448 late pregnancy or around the time of delivery, while preterm births, neonatal and infant death also can
449 belong to this category. For example, exposure to immunomodulating or immunosuppressive
450 substances may increase the risk of infections in the newborn. Other examples include neonatal
451 withdrawal syndrome resulting from sudden discontinuation of exposure to a medicinal product (e.g.
452 opioids), or persistent pulmonary hypertension of the newborn (PPHN), following exposure to
453 serotonergic or vasoconstrictive substances. The likelihood and severity of these effects depend on the
454 pharmacological properties of the substance, the timing, degree and duration of exposure, and the
455 developmental stage at which the fetus is exposed.
456 5.3.1.5. Neurodevelopmental disorders (NDD)
457 The entire nervous system, including the CNS, remains vulnerable to neurotoxic effects throughout
458 pregnancy. Accordingly, in utero exposure to medicinal products may impact on neurodevelopment,
459 even in the absence of structural anomalies. A consensus definition of NDD is not available, however
460 the International Classification of Diseases ICD-11 [16] defines NDD as behavioural and cognitive
461 disorders with onset during the developmental period before 18 years of age, impairing intellectual,
2 Congenital malformation - narrow subset of congenital anomaly - is used for errors in morphogenesis excluding
disruptions or deformations. Birth defect refers to the observable structural anomalies that are present at birth (GVP
Product- or Population-Specific Considerations III: Pregnant and breastfeeding women and their children exposed in utero
or via breastmilk [10])
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 13/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 14 页
462 motor, language, or social functions. Examples include intellectual disability, autism spectrum disorder
463 (ASD), attention-deficit/hyperactivity disorder (ADHD), and learning disorders. These outcomes are
464 important considerations in the long-term safety assessment of medicinal products used during
465 pregnancy.
466 Risk assessment of NDDs following prenatal medicinal product exposure typically relies on human
467 observational studies, as non-clinical models may not fully reflect the complexity of human
468 neurodevelopment. Such studies should use validated measures to assess neurodevelopment, and
469 different types of NDDs should be analysed separately. Combining heterogeneous conditions (e.g.
470 ADHD, ASD, intellectual disability, and learning disorders) into a single composite outcome is
471 problematic because they differ in aetiology, developmental course, and measurement, which can
472 distort risk estimates and obscure clinically meaningful, condition-specific effects. Further, given the
473 potential for delayed diagnosis of exposure-related NDDs, sufficient follow-up periods to detect
474 outcomes that may only emerge later in childhood are needed. The minimum follow-up period should
475 preferably exceed the upper quartile of the median age at diagnosis, i.e. the age by which
476 approximately 75% of affected children have been diagnosed, in the relevant country or setting. Study
477 designs must also account for the multifactorial nature of NDDs and the role of maternal, paternal and
478 child covariates, including relevant environmental factors, in the relationship with the exposure of
479 interest, and control for potential confounders to ensure the reliability and interpretability of results.
480 Consideration as to the role of different methods of ascertainment in geographical or temporal
481 differences in rates of NDD may also be warranted. As a consensus definition for NDD is not available,
482 particular attention should be given to the definition of NDD and its components in any given study or
483 data source.
484 5.3.2. Clinical assessment of data on pregnancy outcomes
485 For clinical assessment of outcomes of exposure to a medicinal product during pregnancy, available
486 sources of evidence are often non-interventional. However, in certain situations, pregnant individuals
487 may be included in clinical trials [1], depending on the medical condition being studied, as well as on
488 the potential risk profile of the substance, or such studies are undertaken with an aim to characterise
489 use of a medicinal product in pregnant individuals. Evaluation of clinical trials follows standard
490 methodology, as appropriate for the specific trial.
491 Given the variation in background prevalence for adverse pregnancy outcomes, and methodologies
492 across data sources and studies, using an appropriate control group from the same data source, as
493 well as similar duration of follow up, is desirable, and should be aimed for, if feasible, to facilitate valid
494 comparisons and a fulsome assessment.
495 For non-interventional pregnancy studies, several common biases need to be considered [12,13].
496 These issues can distort results but can, at least to a certain extent, be addressed through careful
497 study design, appropriate timing definitions, and sensitivity analyses as outlined below:
498 Ascertainment of exposure to the medicinal product in pregnancy:
499 • Spontaneous and solicited reports may be reported prospectively, in which exposure to the
500 medicinal product is reported before the pregnancy outcome is known, either adverse or non-
501 adverse and prior to any prenatal investigation. In contrast, retrospective reports involve cases
502 where exposure is reported after the outcome is already known. While retrospectively reported
503 pregnancies should not be excluded from the assessment, an analysis restricted to prospectively
504 reported pregnancies avoids recall bias.
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 14/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 15 页
505 • Timing and degree of exposure to the medicinal product during pregnancy relative to susceptible
506 periods should be considered. For instance, where pregnancies are exposed late in the first
507 trimester, organogenesis may have already been completed and their capture as exposed
508 pregnancies may bias a harmful effect towards the null. Pharmacokinetics including half-life and
509 placental transfer should be considered in relation to degree of exposure during pregnancy.
510 Furthermore, if exposure occurred prior to conception or the susceptible period for a particular
511 adverse outcome, these parameters have to be considered before excluding exposure to the
512 medicinal product as relevant to the adverse pregnancy outcome of interest.
513 • When comparing adverse pregnancy outcomes following exposure to a medicinal product which
514 occurs after conception with those among pregnancies without such exposure, immortal time bias
515 may occur:
516 − either when the unexposed ‘immortal’ (event-free) time spent prior to exposure is incorrectly
517 classified as exposed, leading to underestimated risk among the exposed pregnancies,
518 − or the unexposed ‘immortal’ time prior to exposure is incorrectly excluded from the analysis,
519 leading to an over-estimated risk among unexposed pregnancies and selection bias
520 (pregnancies included in the exposed cohort had to survive without experiencing the adverse
521 pregnancy outcome in order to be exposed to the medicinal product and therefore may differ
522 from those that were never exposed).
523 This can be avoided by defining cohort entry at treatment initiation or by using a time-varying
524 exposure definition. Alternatively, a landmark design (e.g., gestational week 12 or 20) may be
525 used, although this can introduce live birth bias by excluding early pregnancy losses.
526 Ascertainment of the adverse pregnancy outcome:
527 • The definition of the adverse pregnancy outcome studied, should be considered. Caution is advised
528 in generalising conclusions to subtypes of outcomes or components of a composite outcome not
529 explicitly studied. Population-based studies in real-world data sources may be affected by
530 misclassification, for example due to reliance on a single (diagnostic) code that does not confirm
531 the outcome, overrepresentation of more severe outcomes (e.g. conditions requiring
532 hospitalisation), or underrepresentation of outcomes that may not routinely present to the
533 healthcare system (e.g. early miscarriage, pregnancy termination, or certain minor congenital
534 anomalies).
535 • The period of follow-up may have important implications. Prevalence of major anomalies increases
536 when infants are followed to 12 months of age rather than being evaluated at birth only. In the
537 context of NDDs, assessing CNS function in early childhood present challenges, as some disorders
538 may not manifest until later in childhood, thereby requiring years of follow up to capture
539 adequately.
540 Selection bias:
541 Selection bias in studies of medicinal products exposure during pregnancy can occur when the
542 pregnancies included are not representative of pregnancies in general. This can arise in several ways.
543 For example, left truncation occurs when pregnancies are not observed from conception and need to
544 reach a certain gestational age to be captured in a data source, leading to under-ascertainment of the
545 total number of pregnancies. Live birth bias is another concern and occurs when studies include only
546 live-born infants, potentially underestimating risk if the exposure leads to pregnancy losses that are
547 excluded from analysis. This issue is less relevant in studies of long-term outcomes such as NDDs,
548 which are assessed among live births only. Avoiding selection bias requires careful attention to
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 15/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 16 页
549 inclusion criteria, timing of data capture, and completeness of outcome reporting, especially for early
550 pregnancy losses and non-live births.
551 Confounding bias:
552 • Minimising confounding by indication in pregnancy safety studies requires careful study design,
553 particularly in comparator selection and adjustment for disease-related factors. Restricting the
554 study population to pregnant individuals with the same underlying condition—for example, by
555 including a non-medicated disease comparison group—can help control for confounding at the
556 design stage. However, confounding by disease severity may still remain. To address this,
557 researchers can use an active comparator group (e.g. pregnant individuals treated with alternative
558 treatment for the same indication) and apply statistical adjustment for available markers of disease
559 severity and other confounders to strengthen causal inference. Designs where siblings in
560 unexposed and exposed pregnancies are compared may also be helpful.
561 • Confounding may also arise due to differences between exposed and comparator cohorts in the
562 calendar period over which pregnancies occurred (due to changes in prescribing practices or
563 outcome detection over time) or gestational age over which pregnancies were observed (whereby
564 the cohort observed from an earlier gestational age may be less susceptible to left truncation) and
565 should be addressed.
566 • The potential causal relationship addressed by the research question including relevant maternal,
567 paternal and offspring covariates, and biological (including the indication) as well as environmental
568 factors should be carefully considered. This should include in particular the impact of unmeasured
569 confounding. It is also important to consider over-adjustment bias and collider (selection) bias that
570 could arise due to adjustment for intermediate covariates occurring or measured during or at the
571 end of pregnancy and after the exposure. Evaluation should be informed by clinical knowledge and
572 relevant literature and may be complemented by causal diagrams.
573 Required sample size:
574 There is no predefined number of exposed pregnancies that is sufficient for a reassuring conclusion for
575 any potential risk to the embryo or fetus as this depends on the baseline rate of a particular adverse
576 outcome, and the many factors that influence its detection and reporting. For major congenital
577 malformations (with baseline rates of 3-5%), a few hundred prospectively collected exposed
578 pregnancies, with known outcomes, without a concerning pattern are generally needed to conclude
579 that a manyfold increased risk is less likely.
580 For stronger reassurance, including lack of increased risk for major congenital malformations in specific
581 organs, data of adequate quality from a few thousand prospectively collected exposed pregnancies,
582 with known outcomes not indicating an increased risk are needed.
583 For some adverse pregnancy outcomes, where the background rates are more common than for major
584 congenital malformations, smaller sample sizes may suffice.
585 Observations from different data sources can be pooled provided methodological considerations are
586 considered, for instance use of comparable definitions for the outcomes of interest. The estimated
587 absolute frequency, relative risk or excess risk of an adverse pregnancy outcome following exposure to
588 the medicinal product in utero should in general be presented with a confidence interval, where the
589 width of the interval reflects the precision of the estimate.
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 16/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 17 页
590 6. Assessment regarding breastfeeding
591 6.1. Introduction
592 Clinical data are essential for characterising safety of a medicinal product for a breastfed infant, and to
593 support recommendations regarding treatment of the mother during breastfeeding. Non-clinical data
594 can be complementary, particularly when clinical data are limited. Milk samples collected from
595 lactating 3 or breastfeeding individuals can be used in studies to quantify medicinal product transfer
596 into human milk, allowing calculation of the relative 4 and/or daily 5 infant dose.
597 Pharmacokinetics of the medicinal product, in both mother and infant should also be considered. This
598 does not only cover possible transfer into breast milk, but also systemic exposure in the infant (based
599 on intestinal absorption, distribution, metabolism and elimination). Additionally, the pharmacodynamic
600 properties of the medicinal product should also be considered to assess the potential for adverse
601 reactions in the infant during breastfeeding.
602 Data from follow-up of breastfed infants for potential adverse reactions provides additional valuable
603 insights into the safety of medicinal products during breastfeeding.
604 Modelling and simulation approaches can be used to leverage sparse human data, primarily through
605 population pharmacokinetic modelling. Physiologically based pharmacokinetic modelling (PBPK) needs
606 qualified models, however since the pharmacokinetics of transfer to breast milk is a complex process
607 and the impact of the model will be high, the feasibility of using such models is currently limited.
608 Complementary role of non-clinical data
609 Information regarding excretion into animal milk of the active substance (and/or metabolites) can be
610 of value for the assessment. However, due to interspecies differences in lactation physiology and milk
611 composition, direct quantitative extrapolation of medicinal product concentrations in animal milk to
612 human milk is generally not possible. Non-clinical data alone is generally insufficient to support
613 definitive recommendations in humans, unless the method or model is predictive for human milk
614 transfer or infant exposure/safety in the context of use. In general, recommendations on breastfeeding
615 should be based primarily on human data, while non-clinical data may contribute to risk assessment
616 when adequate human data are lacking, and should be interpreted with caution.
617 6.2. Risk assessment for a breastfed infant
618 Major and complementary factors for risk assessment are described below. Their order below does not
619 reflect their relative importance in the assessment process.
620 6.2.1. Major elements of assessment:
621 • Breastfed infants’ plasma/blood concentrations of the medicinal product, which support direct
622 estimation of systemic exposure.
3
“Lactating” refers to the physiological production of milk, while “breastfeeding” refers to the act of feeding the infant. In
some cases, milk samples may still be collected after breastfeeding has ceased if milk production persists.
4
Relative infant dose (RID) is the estimated daily dose of a medicinal product received by an infant through breast milk
(mg/kg/day), expressed as a percentage of the maternal weight-adjusted dose (mg/kg/day). It provides an estimate of the
extent of medicinal product transfer into breast milk. For quantification of risk for the infant based on RID categories see
section 8.2.1.2.
5 Daily infant dose (DID) is the absolute dose of a medicinal product ingested by an infant through breast milk. It provides a
direct measure of infant medicinal product exposure during breastfeeding. DID can be compared with the therapeutic daily
dose for an infant of the same age and body weight receiving the medicinal product as treatment.
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 17/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 18 页
623 • Estimation of the amount of medicinal product received by infants through breast milk, based on
624 the measured concentration of the active substance /metabolite(s) in human milk, taking into
625 account maternal dose and treatment duration, as well as estimated breastmilk intake by the
626 infant. These could be expressed as a relative infant dose (RID) or a daily infant dose (DID).
627 • Case reports of adverse effects in breastfed infants, which could be signals of potential safety
628 concerns.
629 • Follow up of breastfed infants during their mothers’ treatment, which supports ongoing safety
630 evaluation and could potentially identify both short-term and long-term effects.
631 • Pharmacokinetics in the mother (considering route of administration, plasma concentrations and
632 elimination characteristics), in breast milk and in the infant (particularly gastrointestinal absorption
633 and rate of elimination).
634 • Pharmacodynamic properties and safety profile of the medicinal product, including known class
635 effects and/or data from paediatric use.
636 6.2.2. Complementary data for assessment:
637 As major elements of assessment are usually scarce, all complementary data available should be
638 assessed as well.
639 • The physicochemical properties of the medicinal product, which help estimate the extent of its
640 transfer into breast milk.
641 • Documented use of the medicinal product during breastfeeding, even without formal follow up of
642 breastfed infants.
643 • Infant-specific factors, such as age, overall health status and developmental maturity.
644 • Non-clinical data (e.g., qualified in vitro or in silico methods, transfer of the medicinal product in
645 animal milk, peri- and post-natal or juvenile toxicity studies) which can support estimation of
646 exposure and risk.
647 A pharmacological class effect might not always be relevant for risk assessment for breastfeeding
648 infants, as infant exposure can vary substantially between substances within the same class due to
649 differences in their pharmacokinetic properties, leading to differences in risk classification.
650 Risks from direct dermal exposure to the breastfed child, when there is topical application on the
651 breast or on large skin area should be considered.
652 6.3. Assessment of effect of a medicinal product on lactation (breastmilk
653 production, secretion and composition)
654 Some medicinal products can decrease or increase breast milk production by affecting prolactin levels.
655 Others can affect breast milk secretion by affecting oxytocin levels. Additionally, certain medicinal
656 products may impact the composition or quality of breast milk, or other aspects related to
657 breastfeeding. All existing data on these effects should be considered when relevant.
658 6.4. Proteins including monoclonal antibodies and risk assessment for
659 breastfeeding
660 Physiological processes (including increased paracellular permeability of the mammary epithelium
661 shortly after delivery) result in therapeutic monoclonal antibodies (mAbs) being present in higher
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 18/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 19 页
662 amounts in the colostrum than in mature milk, with rapidly decreasing levels of the mAb thereafter.
663 Gastrointestinal uptake of mAbs by the breastfed infant is also very low, with some possible exception
664 for preterm infants and neonates. Due to intrinsic properties, mAbs are denatured and rendered
665 inactive following acidification in the infant’s stomach, further limiting systemic exposure.
666 A standard recommendation cannot be given for all mAbs, as safety during breastfeeding depends on
667 substance-specific factors including the extent of breast-milk transfer, the infant’s gastrointestinal
668 absorption and the pharmacological mechanism of the mAb. Furthermore, findings from (enhanced)
669 (e)PPND studies, if appropriate (see [2,17,18]), should be considered. However, when the risk profile
670 of the mAb is sufficiently characterised, product-specific recommendations can be made. In some
671 cases, it may be appropriate to restrict breastfeeding only right after birth and to permit breastfeeding,
672 thereafter, based on physiological principles and inherent characteristics of the medicinal product. A
673 similar case-by-case approach may be applicable to other protein-based therapeutics. Standard texts
674 for mAbs or protein-based products with respect to breastfeeding are included in the Annex.
675 7. Integration: From data to product information
676 7.1. General principles
677 The general aim of this section is to guide on how to integrate data from various sources to determine
678 appropriate wording for the SmPC and PL in relation to male and female fertility, for females of
679 reproductive potential, and for pregnant or breastfeeding individuals. Efforts should be made to
680 provide the reasons for the recommendations given. This information is important for the health care
681 professional and for the (to be) treated individual to support an informed decision.
682 In the overall assessment, available knowledge from all relevant data sources should be considered.
683 This includes clinical experience, pharmacological activity, pharmacokinetic data, results from non-
684 clinical studies, and knowledge about medicinal products within the same class; based on information
685 from e.g. clinical studies (interventional or non-interventional), non-clinical testing (in vitro or in vivo),
686 spontaneous reporting as well as literature. Aspects such as alternative treatments, impact of the
687 maternal disease on the pregnancy, and clinical practices should also be considered. The more
688 extensive clinical data being available, the less emphasis is put on non-clinical studies in the
689 integration process.
690 Within each specific area, following an integrated assessment of available evidence and its strengths
691 and limitations, a general risk classification is concluded upon, as follows.
692 • A risk is considered identified if there is sufficient scientific evidence from non-clinical and/or
693 clinical data, and in some instances taking pharmacodynamic effects into account, to conclude,
694 with at least a reasonable possibility, that the adverse outcome is caused by the medicinal
695 product.
696 • A risk is considered potential if scientific evidence from non-clinical and/or clinical data, and in
697 some instances taking pharmacodynamic effects into account, suggests a possible causal
698 relationship between the adverse outcome and the medicinal product. It can also correspond to
699 insufficient non-clinical and/or clinical data, but where available data raise concerns.
700 • A risk is considered unlikely when a sufficiently large amount of non-clinical and / or clinical data
701 of adequate quality, including knowledge of the pharmacodynamic effects, do not raise concerns
702 for adverse outcomes.
703 • Available data do not raise specific concerns but are insufficient to draw final conclusions.
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 19/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 20 页
704 This overall assessment guides the information, and recommendations, including risk minimisation
705 measures, to be included into the SmPC and PL. The risk classification can also guide risk management
706 planning, as outlined in GVP Module V on Risk management systems [19]. Some guiding elements
707 regarding the risk classification within each specific area are presented below.
708 The PI, and if appropriate, the recommendations for use in these different populations, should be
709 updated when relevant new data become available.
710 7.1.1. Fertility
711 The evaluation of effects on human fertility mostly relies on the outcome of non-clinical studies and
712 knowledge of any effect that could be related to the mechanism of action. If available, clinical data
713 should be integrated in the assessment. Regarding non-clinical studies, the assessment should be
714 based on the results of genotoxicity studies, FEED studies and/or repeat-dose toxicity studies. The
715 latter generally provides important information contributing to further characterise potential effects on
716 fertility based on pathology and functional parameters for both male and female reproductive function.
717 Evidence of impaired fertility includes abnormal sperm parameters, menstrual or ovulatory
718 disturbances, testicular or ovarian toxicity, reduced conception rates in animals, and additional
719 mechanistic evidence regarding reproductive hormones or germ cell function.
720 For the assessment of the risk classification, some guiding elements are outlined below:
721 7.1.1.1. Identified risk for impaired fertility
722 • Non-clinical data of human relevance and of sufficient quality, and / or clinical data of sufficient
723 quality show a risk for impaired fertility.
724 • A plausible mechanism of action has been identified that leads to risk for impaired fertility.
725 7.1.1.2. Potential risk for impaired fertility
726 • Non-clinical data suggest adverse effects on fertility, but human relevance is uncertain, and there
727 is no or insufficient clinical evidence to conclude causality.
728 7.1.1.3. Unlikely risk for impaired fertility
729 • Non-clinical data do not show or suggest impaired fertility and there is no clinical evidence
730 suggesting adverse effects on fertility.
731 7.1.1.4. Insufficient data with respect to fertility
732 • Available data do not raise specific concerns but are insufficient to draw final conclusions e.g. non-
733 clinical findings are inconclusive, and clinical evidence is insufficient to fully assess risk of adverse
734 effects on fertility.
735 7.1.2. Pregnancy
736 An integrated approach for risk assessment of prenatal exposure and adverse pregnancy outcomes is
737 presented below, in the form of questions adapted from teratological principles [20-22], modern
738 epidemiological reasoning and principles for assessment of medicinal products safety in pregnancy. The
739 order of questions below does not reflect the relative importance of these in the assessment process.
740 While the reliability of the overall assessment will depend on the extent to which clear and definitive
741 answers to these questions can be obtained, in practice conclusions may need to be drawn
742 acknowledging that at the time of the assessment some of these questions remain unanswered.
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 20/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 21 页
743 7.1.2.1. Integrated questions for risk assessment of causation between prenatal exposure
744 and adverse pregnancy outcomes
745 1. Exposure Timing: Was there documented exposure to the medicinal product at critical times
746 during prenatal development, and does the exposure precede the observed adverse pregnancy
747 outcomes? Is there (animal) data on placental transfer?
748 2. Consistent Findings: Are there consistent findings by multiple epidemiological studies of
749 adequate quality showing a relationship between exposure and adverse pregnancy outcomes? Do other
750 data sources provide further support for concern?
751 3. Strength of Association: What is the magnitude of the observed association between
752 exposure and adverse pregnancy outcomes as indicated by clinical studies (for a non-interventional
753 study, consider the degree of potential residual confounding)? What is the precision of the estimated
754 association?
755 4. Specificity of Clinical Phenotype: Is there a specific pattern of defects reported after the
756 exposure? An embryo/fetotoxic substance rarely causes congenital malformations in a wide range of
757 organs. A specific pattern of congenital malformations generally increases the likelihood of a causal
758 relationship.
759 5. Biological Mechanism: Does the association make biological sense, supported by evidence of
760 a plausible biological mechanism of action, from non-clinical and / or clinical data?
761 6. Class effects (i.e. analogy): Have similar medicinal products shown similar adverse
762 outcomes, pointing to a class effect that supports a causal relationship, based on non-clinical and / or
763 clinical data?
764 7. Dose-Response: Is there evidence of a dose-response relationship, indicating that higher
765 exposure levels are associated with greater risks of adverse outcomes?
766 8. Non-clinical Evidence: Is there evidence of adverse embryo-fetal effects in toxicity studies
767 supporting causality between exposure and adverse pregnancy outcomes? Is the substance genotoxic?
768 9. Integration and Coherence: Does the overall body of evidence support a coherent and
769 consistent conclusion about a causal relationship between the exposure and adverse pregnancy
770 outcomes?
771 7.1.2.2. Identified risk for adverse pregnancy outcomes
772 The evidence supporting an identified risk may include:
773 • Study(ies) in humans, with designs and results of sufficient quality to conclude on causality,
774 consistently suggesting an increase in the overall frequency of an adverse pregnancy outcome (or
775 a specific type of an adverse pregnancy outcome). Factors that may be considered include the
776 strength of the association of exposure to the medicinal product in utero with the adverse
777 pregnancy outcome, any evidence of a dose-response relationship, and coherence of the overall
778 body of evidence, as described in the questions in section 7.1.2.1 above.
779 • Several relevant case-reports, where assessment conclude that a causal relationship between the
780 medicinal product and the adverse pregnancy outcome is at least a reasonable possibility. Factors
781 supporting such relationship include known exposure during the critical period, low frequency of
782 the adverse pregnancy outcome in the general population, specific effect (e.g. an unusual pattern
783 of an adverse pregnancy outcome ), plausible mechanism for the adverse pregnancy outcome in
784 relation to the mechanism of action, known association of other medicinal products in the same
785 class with the adverse pregnancy outcome, and consistency with non-clinical findings of a harmful
786 effect of exposure to the medicinal product in pregnancy.
787 • Non-clinical data which show a harmful effect(s) after exposure to the medicinal product during
788 gestation, with suspected relevance to humans and with insufficient clinical data;
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 21/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 22 页
789 • Mechanism of action with evidence of being a risk for adverse pregnancy outcomes, including
790 known class effects.
791 7.1.2.3. Potential risk for adverse pregnancy outcomes
792 The evidence supporting a potential risk may include:
793 • Non-clinical data which raise concerns for a harmful effect(s) due to exposure to the medicinal
794 product during gestation, with possible relevance to humans but with insufficient clinical data.
795 • Mechanism of action which is suspected to adversely interfere with development. This includes
796 possible class effects of risk for adverse pregnancy outcome(s).
797 and/or
798 • Studies or sources of data such as analysis of human case reports suggesting an increase in the
799 overall frequency of the adverse pregnancy outcome, but insufficient evidence to conclude on
800 causality, due to limitations, inconsistency of findings across data sources, or a lack of other
801 studies or data to corroborate the findings of a single data source. This includes the relevance of
802 the methodology and results of each study and source of data.
803 7.1.2.4. Adverse effects unlikely in humans
804 When a sufficiently large amount of clinical data of adequate quality is available, not raising concerns
805 for adverse pregnancy outcomes, and when knowledge regarding the pharmacological effects of the
806 substance does not raise such concern, it can be concluded that adverse pregnancy effects are
807 unlikely. This would generally correspond to several thousand pregnancies, well documented within a
808 relevant study setting. In this situation, beyond pharmacological data, the non-clinical data have less
809 importance for risk assessment. For some formulations, pharmacokinetic data in humans, showing
810 very low systemic exposure can also support risk assessment.
811 7.1.2.5. Insufficient data with respect to pregnancy
812 For substances where non-clinical data, including knowledge of pharmacological effects, do not raise
813 concerns, but where clinical data are limited, and therefore no firm reassuring conclusions can be
814 drawn. In these situations, available data are considered insufficient.
815 7.1.3. Breastfeeding
816 Following the integrated assessment outlined in section 6, a general risk classification can be applied,
817 based on the available evidence and its strengths and limitations. In all situations below, if the
818 medicinal product is not absorbed from the infant’s gastrointestinal tract, it will not cause systemic
819 effects, but it can still cause local effects in the gastrointestinal tract, such as diarrhoea.
820 7.1.3.1. Identified risk for breastfed infants
821 The evidence supporting an identified risk may include:
822 • Adverse reactions in breastfed infants in line with the known safety profile and/or the
823 pharmacodynamic effects of the substance.
824 • Clinically relevant systemic exposure of breastfed infants to the active substance (and/or its
825 metabolites), and/or if the substance (and/or metabolites) is present in significant amounts in
826 breast milk can also support concluding on an identified risk for breastfed infants.
827 7.1.3.2. Potential risk for breastfed infants
828 The evidence supporting a potential risk may include:
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 22/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 23 页
829 • Safety profile of a medicinal product including relevant risks, taking pharmacodynamic properties
830 as well as class effects into account, but data on transfer into breastmilk or data on exposure of
831 breastfed infants are lacking. This assessment can in some instances be based on non-clinical data
832 solely, particularly from toxicity testing including peri- and post-natal or juvenile toxicity studies.
833 • Limited data on milk transfer or of infant exposure suggest levels of the active substance, that may
834 cause adverse effects in the infant.
835 • Data on milk transfer or infant exposure is not available or limited, but physicochemical properties
836 of active substance, data from a qualified physiologically based pharmacokinetic model using
837 scarce human samples, or from a qualified in vitro method, point to possible high transfer into
838 breast milk.
839 7.1.3.3. Adverse effects unlikely in breastfed infants
840 The evidence supporting that adverse effects are unlikely in breastfed infant may include:
841 • Sufficient human data from the use in breastfeeding individuals not suggesting a risk, including no
842 reported adverse reactions in the infant.
843 • Robust data showing that the levels of the substance/metabolites in breastmilk are so low (see
844 section 8.2.1.2) that they are unlikely to pose a concern to the breastfed infant. Alternatively, data
845 may show that the medicinal product does not enter breast milk and/or is not absorbed
846 systemically by the infant.
847 7.1.3.4. Insufficient data with respect to breastfeeding
848 When clinical data from breastfed infants, or on levels in breastmilk are absent or limited, data are
849 considered insufficient. The overall safety profile of the substance does not raise serious concerns.
850 7.2. Principles for recommendations
851 7.2.1. Fertility
852 In most cases, recommendations are not provided in the PI in relation to data on fertility. However, on
853 a case-by-case basis, it may be warranted to include a recommendation into the PI, taking the following
854 into account (see section 8.2.1.3):
855 • In case of an identified risk of temporarily impaired fertility, a recommendation not to use the
856 medicinal product in males and/or females attempting to conceive, can be considered. Further, in
857 individuals with difficulties conceiving or who are undergoing investigation of infertility, a
858 recommendation to withdraw the treatment can be considered.
859 • If a risk of irreversible damage of male and/or female gametes has been identified, inclusion of a
860 recommendation seeking counselling for sperm or oocyte preservation before starting treatment,
861 can be considered.
862 7.2.2. Pregnancy
863 The potential risks of in utero exposure to the medicinal product for the developing embryo or fetus
864 (see section 7.1.2), the potential benefits of treatment for the pregnant individual, the risks the
865 untreated condition may pose to both, and alternative treatments, should be taken into account when
866 determining the recommendation for use of the medicinal product in pregnancy. These
867 recommendations should be based on available information from:
868 • Clinical experience
869 • Relevant non-clinical studies
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 23/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 24 页
870 • Need for treatment
871 7.2.2.1. Clinical experience
872 In case of a potential or an identified risk for serious adverse pregnancy outcomes such as major
873 congenital malformations or NDD (see section 7.1.2), a contraindication to use of the medicinal
874 product in pregnancy, or during defined parts of pregnancy should be considered.
875 A contraindication should also be considered if the medicinal product has a pharmacological profile
876 (mechanism of action) similar to that of a substance or therapeutic class where serious adverse
877 pregnancy outcomes are an identified risk in humans.
878 In cases when available non-clinical and clinical evidence suggests a lower concern for serious adverse
879 pregnancy outcomes, less strict recommendations should be given, taking the need for treatment into
880 account (see section 7.2.2.3).
881 7.2.2.2. Relevant non-clinical studies
882 In some circumstances, a contraindication in humans may be based on non-clinical data only, after a
883 comprehensive assessment of all relevant aspects, such as:
884 • Pharmacology/mechanism of action: properties known to cause direct or indirect embryo-fetal
885 damage
886 • Genotoxic potential
887 • EFD: lethality, congenital malformations, embryo-fetal toxicity including growth retardation, and
888 functional changes, taking the severity of the observed findings, and the exposure level in relation
889 to the human therapeutic exposure into account.
890 7.2.2.3. Need for treatment
891 Where the assessment concludes on an identified or potential risk for adverse pregnancy outcomes,
892 such risk has to be weighed against the potential benefit of the treatment for the pregnant individual,
893 before a recommendation is concluded upon. The following points should be taken into consideration:
894 • Serious or life-threatening disease, as well as the impact of pregnancy on the disease and the
895 impact of the disease on pregnancy.
896 • An alternative treatment option, with relevant therapeutic benefit, is available that is safer when
897 used in pregnancy.
898 • Treatment can be modified, postponed or avoided.
899 A contraindication in pregnancy should only be considered if either, a) safer alternative treatment
900 option is available or, b) treatment can be postponed until after pregnancy.
901 If there is no safer treatment option available and treatment cannot be delayed, the medicinal product
902 should generally not be contraindicated in pregnancy. However, strict recommendations against use in
903 pregnancy and actions to take in case of pregnancy occurring, or if pregnancy is being planned while
904 on treatment, should be included in the SmPC and PL. See also section 8.2.1.
905 7.2.2.4. Recommendations for use during pregnancy
906 Recommendations for use in pregnancy should be based on all relevant evidence, and take the risk
907 classification described in section 7.1.2 into account:
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 24/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 25 页
908 • Medicinal product is contraindicated during pregnancy – This applies when an identified or
909 potential risk for serious adverse pregnancy outcome(s) has been concluded upon, and when
910 treatment can be postponed until after pregnancy, or the medical condition can be managed by an
911 alternative treatment option.
912 • Medicinal product should not be used during pregnancy unless the medical condition
913 requires treatment with {active substance} and cannot be managed by an alternative
914 treatment – This applies when an identified or potential risk for a serious adverse pregnancy
915 outcome(s) is concluded upon, and treatment cannot be postponed until after pregnancy or the
916 medical condition cannot be managed by an alternative treatment option. This also applies when
917 available data are insufficient to draw conclusions about the risk of in utero exposure, but available
918 evidence raises concerns for adverse pregnancy outcomes. This advice can be adapted to meet
919 specific exposure windows when adequately justified.
920 • Use of the medicinal product may be considered during pregnancy – This applies when data
921 are limited but available evidence does not raise concerns for serious pregnancy outcomes.
922 • Medicinal product can be used during pregnancy – This applies when the risk assessment of
923 available evidence concludes on an unlikely risk.
924 7.2.3. Individuals of reproductive potential
925 The need for a recommendation for females of reproductive potential to use effective contraception
926 [8,23] to prevent unplanned pregnancy during treatment should be considered when a medicinal
927 product has an identified or potential risk for adverse pregnancy outcomes.
928 Factors that should be taken into account include the robustness of available evidence, the magnitude
929 and severity of identified or potential risk, the target population, and availability of alternative
930 treatment options. Recommendations for females of reproductive potential will also depend on the
931 recommendation with respect to use of the medicinal product in pregnancy.
932 In exceptional cases, for an active substance which can cause serious adverse pregnancy outcomes in
933 human, there can be concerns that the substance can affect embryo-fetal development through
934 transfer to a pregnant individual via seminal fluid from a male individual being treated with the
935 substance [8]. In such situation, a recommendation for male individuals to use a condom if engaged in
936 sexual activity with a pregnant individual or a female of reproductive potential not using effective
937 contraception may be considered.
938 7.2.3.1. Recommendations for use in females of reproductive potential
939 Where the medicinal product is contraindicated in pregnancy (see 7.2.2.4):
940 • In exceptional circumstances, if a risk of serious embryo-fetal or neonatal harm of larger absolute
941 magnitude has been concluded, the need to prevent exposure to the medicinal product in
942 pregnancy by contraindicating use in females of reproductive potential, unless they are using
943 effective contraception, should be considered.
944 • In most situations, such contraindication is not warranted. Instead, a recommendation that the
945 medicinal product should not be used in females of reproductive potential unless they are using
946 effective contraception is sufficient.
947 • In some situations, depending on the indication and the context for use e.g. acute treatments, a
948 recommendation to use effective contraception in females of reproductive potential is not
949 warranted.
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 25/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 26 页
950 Where the medicinal product should not be used during pregnancy unless the medical condition
951 requires treatment with the active substance and cannot be managed by an alternative treatment (see
952 7.2.2.4):
953 • When an identified or potential risk for a serious adverse pregnancy outcome(s) is concluded upon,
954 or when available data are insufficient to draw conclusions, but available evidence raises concerns
955 for adverse pregnancy outcomes, a recommendation that females of reproductive potential should
956 use effective contraception should be considered, depending on the potential safety concerns.
957 For all the above situations, there may be circumstances where the possibility for becoming pregnant is
958 very low for a female of reproductive potential. In these circumstances an effective contraception may
959 not be necessary, following discussion with the prescriber. Where such circumstances may change, this
960 should be reviewed regularly.
961 7.2.3.2. Recommendations on risk minimisation measures
962 When the assessment has concluded on an identified risk or a potential risk of serious adverse
963 pregnancy outcomes, concrete measures to avoid or minimise exposure to the medicinal product
964 during pregnancy should be provided in the PI (see section 8 and Annex).
965 Actions to ensure adherence to recommendations in the PI can be combined with additional tools of
966 risk minimisation, including, in exceptional cases, a pregnancy prevention programme (See GVP
967 Module XVI Addendum I – Risk minimisation measures for medicinal products with embryo-fetal risks
968 [24]).
969 7.2.4. Breastfeeding
970 Recommendations for use during breastfeeding should be based on all relevant evidence, including the
971 known benefits of breastfeeding, and take the risk classification described in section 7.1.3 into account.
972 7.2.4.1. Considerations for recommendations in breastfeeding
973 A medicinal product should only be contraindicated during breastfeeding if there exists an identified or
974 potential risk of serious adverse reaction(s) in the breastfed infant. In some circumstances, a potential
975 serious risk can be based on non-clinical or indirect evidence only. Absence of relevant data on
976 breastfeeding alone is not a sufficient reason for a contraindication. The reasons for the (potential or
977 identified) serious risk should be described. If there is concern about a possibility of a potential serious
978 risk for the infant, a recommendation that the medicinal product should not be used during
979 breastfeeding is warranted.
980 If there is a concern about an identified or potential non-serious risk for the infant, a cautionary
981 warning about the use of the medicinal product during breastfeeding should be included, with an
982 explanation provided. This can include that use of the medicinal product during breastfeeding may be
983 considered, with a recommendation to closely observe the breastfed infant (indicating which adverse
984 reactions could be expected) to determine when discontinuation or an interruption of breastfeeding
985 becomes necessary. Alternatively, recommendations on temporary interruption of breastfeeding
986 should be provided, if this is a feasible alternative instead of recommending stopping breastfeeding. In
987 such cases, the duration of interruption should be stated, and the time interval should be estimated
988 based on the half-life of the active substance (and/or active metabolites), oral bioavailability, and
989 safety profile. As a general rule, active substance (and/or metabolites) levels in breast milk are
990 considered negligible after approximately five elimination half-lives [25].
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 26/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 27 页
991 7.2.4.2. Recommendations for use during breastfeeding
992 Recommendations for use during breastfeeding should be based on all relevant evidence, and take the
993 risk classification described in section 7.1.3 into account:
994 • Medicinal product is contraindicated during breastfeeding– this applies in case of an
995 identified serious risk or a potential serious risk for the breastfed infant.
996 • Medicinal product should not be used during breastfeeding– this applies when serious risk
997 for breastfed infant has not been demonstrated but is a possibility.
998 • Medicinal product may be considered during breastfeeding– this applies when data are limited
999 but available evidence does not raise concerns for serious risk. This includes infant exposure being
1000 unknown or may potentially lead to adverse effects, although a serious risk is not expected.
1001 • Medicinal product can be used during breastfeeding– this applies when the risk assessment
1002 of available evidence concludes on an unlikely risk.
1003 8. Reflections for the product information
1004 8.1. Introduction
1005 In the SmPC and PL, information regarding male and female fertility, pregnancy and breastfeeding
1006 should include different elements namely:
1007 • Recommendations with respect to use of the medicinal product in these different populations, as
1008 appropriate.
1009 • A concise summary of available data concerning use of the medicinal product in these different
1010 populations, including potential and identified risks, impact on relevant outcomes and areas of
1011 uncertainty.
1012 • Details of recommended actions for risk minimisation to prevent or minimise exposure during
1013 pregnancy, or during breastfeeding, as appropriate, and / or recommendations for females of
1014 reproductive potential, or in some cases, in relation to male fertility.
1015 The overall aim of the wording in the PI is to enable informed and shared decision-making by
1016 healthcare professionals and patients (or individuals and facilitate adherence to risk minimisation
1017 measures.
1018 8.2. Summary of product characteristics (SmPC)
1019 In section 4.6 Fertility, pregnancy and lactation, information and recommendations regarding use in
1020 pregnancy, and breastfeeding, as well as information regarding fertility should always be included.
1021 In some cases, other sections, e.g. section 4.3 Contraindications, 4.4 Warnings and precautions, 4.5
1022 Interactions and section 4.8 Undesirable effects may also complement section 4.6.
1023 Relevant pharmacokinetic data should be reflected in section 5.2 and non-clinical data should be
1024 reflected in section 5.3, including addressing reproductive and developmental effects. In all relevant
1025 cases, cross-references between sections should be added.
1026 If additional risk minimisation tools are implemented (e.g. educational materials for health care
1027 professionals and/or patients), these should generally be referred to in section 4.4 or 4.6 of the SmPC,
1028 and in section 2 of the PL.
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 27/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 28 页
1029 8.2.1. Section 4.6 of the SmPC
1030 8.2.1.1. Subsection on pregnancy and females of reproductive potential
1031 Clinical data:
1032 Describe the extent of human experience and from which data sources (see sections 5.3 and 7.1.2).
1033 When at least a few hundred prospectively observed exposed pregnancies with known outcomes are
1034 available, it is generally appropriate to detail the numbers of pregnancies, particularly when there are
1035 data from relevant clinical studies (clinical trials or pharmacoepidemiological studies). Where the
1036 number of exposed pregnancies with known outcomes is small (generally less than 100), a more
1037 general description of the data, e.g. being ‘limited’ is more suitable, to reflect data limitations. It will
1038 also avoid small specific numbers becoming rapidly outdated.
1039 If data are sufficiently robust, e.g. results from clinical trials or pharmacoepidemiological studies of
1040 adequate size and quality, the knowledge on main adverse pregnancy outcomes should be
1041 summarised. In case there is a reasonable possibility for a causal relationship between an adverse
1042 pregnancy outcome and exposure, describe:
1043 • Primary sources of evidence for the observed risk, including (where available) details on pregnancy
1044 exposure (timing, duration), the magnitude of the risk (relative, absolute, and background
1045 frequencies), and the nature of the risk (such as patterns of congenital malformations or NDD and
1046 their severity). Include the comparator used in relevant studies, if applicable.
1047 • Relevant areas of uncertainty regarding adverse pregnancy outcomes. This may in some
1048 circumstances relate to studies with different results, leading to the overall picture being
1049 inconclusive.
1050 Also, when adequate clinical data provide reassurance regarding use in pregnancy, the primary sources
1051 of evidence, including details on pregnancy exposure (timing, duration), main pregnancy outcomes,
1052 and the comparator used if applicable, should be described. If some uncertainties remain, those should
1053 be summarised shortly.
1054 Non-clinical data:
1055 Only general conclusions of the non-clinical evaluation should be included in section 4.6 of the SmPC.
1056 Further details should be provided in section 5.3 of the SmPC.
1057 Recommendations:
1058 As detailed in sections 7.2.2. and 7.2.3., a recommendation should be included, which generally
1059 ranges from a contraindication in pregnancy and in females of reproductive potential not using
1060 effective contraception, to can be used in pregnancy.
1061 In case measures to minimise or avoid exposure during pregnancy are warranted, those should be
1062 described (see GVP Module XVI Addendum I [24] for further details), including:
1063 • Risk counselling including the potential risks of using the medicinal product during pregnancy
1064 and the recommendations regarding use during pregnancy.
1065 • Pregnancy testing to ensure excluding pregnancy before initiating treatment with the medicinal
1066 product (and in exceptional situations during and after treatment discontinuation). Pregnancy
1067 testing would generally be included in the PI as a risk minimisation measure when use in
1068 pregnancy is contraindicated.
1069 • When effective contraception is recommended, the exact type of contraceptive method should
1070 not be defined, since selection of a contraceptive method should be evaluated individually by the
1071 prescriber and person to be treated. However, in some instances, a medicinal product may impair
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 28/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 29 页
1072 the effectiveness of some methods of contraception. Such data should be described in the product
1073 information (e.g. section 4.5) and may lead to recommendations on type of contraceptive
1074 methods. Based on the pharmacokinetic properties of the medicinal products, duration of
1075 contraceptive use after stopping treatment should be included in the PI.
1076 • Actions if a patient considers a pregnancy during treatment or in the event of an unplanned
1077 pregnancy during treatment with the medicinal product. This includes promptly contacting a
1078 healthcare professional to consider the possibility of e.g. interrupting treatment, dose reduction,
1079 switching to an alternative treatment and/ or relevant specific monitoring during pregnancy such
1080 as fetal ultrasound or of the neonate/infant.
1081 • Consideration to add advice that treatment should be supervised by a physician who is experienced
1082 in the management of the medical condition and the use of the medicinal product, and on regular
1083 medication reviews.
1084 • Where relevant, consider advice to deter patients from donating blood during and for a specified
1085 period after use of the medicinal product, and/or to deter patients from donating semen during and
1086 for a specified period after use of the medicinal product.
1087 8.2.1.2. Subsection on breastfeeding
1088 All relevant data which are the basis for the recommendation should be shortly described (see also
1089 sections 7.1.3 and 7.2.4). This includes conclusions of pharmacokinetic studies e.g. plasma
1090 concentrations in breastfed infants, concentrations of the active substance and/or its metabolite(s) in
1091 human breast milk (including if it is unknown), calculated relative or daily infant dose. If non-clinical
1092 data are part of the basis for the recommendation, these should be shortly summarised, with reference
1093 to section 5.3 for further details. Estimated absorption from an infant´s gastrointestinal tract should be
1094 reported if relevant.
1095 If possible, quantify the extent of transfer into breastmilk using relative infant dose categories (<2%
1096 minimal, 2–5% low, 5–10% moderate, >10% high) [25] and specify relevant details such as route of
1097 administration and/or dosage strength. If known, the amount relative to a therapeutic infant dose and
1098 infant plasma/blood levels could be mentioned.
1099 Information on known risks in breastfed infants, or long-term clinical experience without observed risks
1100 of specific serious concern, should be included if available.
1101 Other important information should be mentioned, such as large size of molecules possibly precluding
1102 breast milk transfer, lack of oral bioavailability, reminder of safety profile, among others.
1103 The SmPC should describe, when applicable, measures that are intended to minimise exposure of the
1104 breastfed child to the medicinal product and/or its active metabolite(s), such as timing the
1105 administration of the medicinal product relative to breastfeeding and/or short-term interruption of
1106 breastfeeding.
1107 8.2.1.3. Subsection on fertility
1108 The SmPC should describe potential impacts on fertility whenever available evidence indicates that the
1109 medicinal product may impair fertility, including whether temporary or irreversible effects.
1110 Main information on the possible effects of the medicinal product on fertility should be included:
1111 • Clinical data if available.
1112 • Relevant conclusions from non-clinical toxicity studies if available. Further details should be
1113 included in Section 5.3 of the SmPC.
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 29/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 30 页
1114 On a case-by-case basis (see sections 7.1.1 and 7.2.1), it may also be warranted to include advice on
1115 practical considerations e.g., avoiding use in the context of assisted reproductive technology
1116 treatment, for individuals with difficulties conceiving or who are undergoing investigation of infertility,
1117 counselling for gamete preservation or advice not to donate gametes.
1118 Cross-references could be included to other relevant sections of the PI (e.g. section 4.4 of the SmPC),
1119 if appropriate.
1120 8.2.2. Section 4.3 of the SmPC
1121 • Contraindication in pregnancy (see section 7.2.2.4), or defined parts of pregnancy, with cross-
1122 references to section 4.4 (if considered appropriate) and section 4.6 where further information
1123 about the background to the contraindication as well as all risk minimisation measures should be
1124 provided.
1125 • Contraindication in females of reproductive potential (see section 7.2.3.1) unless they are using
1126 effective contraception, with cross-references to section 4.4, if considered appropriate and to
1127 section 4.6 where further information about the background to the contraindication should be
1128 provided.
1129 • Contraindication in breastfeeding (see section 7.2.4.2), with cross-references to section 4.6 where
1130 reasons for contraindication should be explained.
1131 8.2.3. Section 4.4 of the SmPC
1132 If a medicinal product is contraindicated in pregnancy and/ or in females of reproductive potential,
1133 including if a pregnancy prevention programme is implemented (see GVP Module XVI Addendum I
1134 [24]), section 4.4 may shortly state the key risk(s) justifying a contraindication, and shortly focus on
1135 key measures to minimise risk, including pregnancy testing before treatment initiation and use of
1136 contraception, with cross-reference to section 4.3, 4.5 (if applicable) and 4.6, where further details on
1137 the risk as well as on risk minimisation measures should be described.
1138 8.2.4. Section 4.5 of the SmPC
1139 If contraceptive measures are recommended for females of reproductive potential (see section 7.2.3),
1140 relevant information concerning interactions of the medicinal product with methods of contraception
1141 should be addressed, with cross-references to e.g. sections 4.3, 4.4, 4.6 and 5.2.
1142 8.3. Package leaflet (PL)
1143 The package leaflet should reflect all relevant information from the SmPC, and be written in a
1144 comprehensible, lay-person friendly language.
1145 Section 2
1146 Contraindications and precautions for use or special warnings should be included in Section 2 of the PL.
1147 Under Do not take [X]:
1148 • Contraindication if the patient (or individual) is pregnant (if considered appropriate)
1149 • Contraindication if the patient (or individual) is a female able to have a child (if considered
1150 appropriate) unless using effective birth control.
1151 • Contraindication if the patient (or individual) is breastfeeding (if considered appropriate)
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 30/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 31 页
1152 Under Pregnancy, breastfeeding and fertility:
1153 • Contraindications or recommendations with respect to use of the medicinal product in pregnant
1154 patients (or individuals), during breastfeeding or in female patients (or individuals who can have a
1155 child.
1156 • Information on the risks following exposure to the medicinal product in pregnancy, including
1157 (where available):
1158 − The type of adverse pregnancy outcome, e.g., the pattern of congenital malformations or the
1159 specific type of NDD associated with use of the medicinal product during pregnancy, including
1160 severity.
1161 − The magnitude of the risk(s) preferably described in absolute terms and in relation to a
1162 relevant comparator. For instance, expressing the number of babies being affected by the
1163 adverse effect among every 100 babies born to individuals taking the medicinal product,
1164 compared with the same figures for babies born to individuals not taking the medicinal product
1165 or for babies born of individuals who do not have the indication for the medicinal product (as
1166 appropriate).
1167 • If available, information in relation to dose or duration of use.
1168 • Lack of evidence, and/or reassuring findings, if appropriate.
1169 • Information on actions to take to prevent exposure to the medicinal product during pregnancy,
1170 during breastfeeding and in an individual who can have a child, in line with recommendations in
1171 the SmPC. Examples on actions to take can be given in the PL as aid to the individual.
1172 • Information on actions if pregnancy occurs or is considered during treatment, as appropriate, in
1173 line with recommendations in the SmPC.
1174 • Information on the possible effects following exposure to the medicinal product on male and female
1175 fertility.
1176
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 31/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 32 页
1177 9. References
1178 1. ICH E21 Guideline on inclusion of pregnant and breastfeeding individuals in clinical trials
1179 (EMA/CHMP/ICH/149462/2025)
1180 2. ICH S5 (R3) Guideline on detection of reproductive and developmental toxicity for human
1181 pharmaceuticals (EMA/CHMP/ICH/544278/1998)
1182 3. ICH S3A Toxicokinetics: A Guidance for Assessing Systemic Exposure in Toxicology Studies
1183 (CPMP/ICH/384/95 (ICH S3A)).
1184 4. ICH M3(R2) Guideline on non-clinical safety studies for the conduct of human clinical trials and
1185 marketing authorisation for pharmaceuticals (EMA/CPMP/ICH/286/1995)
1186 5. ICH M3 (R2) - questions and answers (EMA/CHMP/ICH/507008/2011)
1187 6. Guideline on non-clinical testing for inadvertent germline transmission of gene transfer vectors
1188 (EMEA/273974/2005)
1189 7. ICH S2 (R1) Guideline on genotoxicity testing and data interpretation for pharmaceuticals intended
1190 for human use (Step 5. 2013) (EMA/CHMP/ICH/126642/2008)
1191 8. Safety Working Party (SWP) / Non-clinical Working Party (NcWP) recommendations on the duration
1192 of contraception following the end of treatment with a genotoxic drug
1193 (EMA/CHMP/SWP/74077/2020)
1194 9. WHO. Guideline for the prevention, diagnosis and treatment of infertility: summary of
1195 recommendations. World Health Organization. (2025).
1196 10. Guideline on good pharmacovigilance practices (GVP): Product- or population-specific
1197 considerations III: Pregnant and breastfeeding women and their children exposed in utero or via
1198 breastmilk (EMA/653036/2019)
1199 11. Euro-Peristat. Live births and stillbirths annual summary statistics 2022, Office for National
1200 Statistics. Available at https://www.europeristat.com/wp-content/uploads/2022/11/Euro-
1201 Peristat_Fact_sheets_2022_for_upload.pdf
1202 12. The European Network of Centres for Pharmacoepidemiology and Pharmacovigilance (ENCePP)
1203 Guide on Methodological Standards in Pharmacoepidemiology. Available at The European Network
1204 of Centres for Pharmacoepidemiology and Pharmacovigilance (ENCePP) Guide on Methodological
1205 Standards in Pharmacoepidemiology - Revision 11
1206 13. ENCePP. Annex 2 to the Guide on Methodological Standards in Pharmacoepidemiology Guidance on
1207 methods for the evaluation of medicines in pregnancy and breastfeeding. Available at Annex 2 to
1208 the Guide on Methodological Standards in Pharmacoepidemiology - Guidance on methods for the
1209 evaluation of medicines in pregnancy and breastfeeding (2nd Edition) (europa.eu)
1210 14. EUROCAT. European Platform on Rare Disease Registration (EU-RD-Platform), European
1211 Commission. Available at https://eu-rd-platform.jrc.ec.europa.eu/eurocat/eurocat-network/jrc-
1212 eurocat-central-registry_en
1213 15. Birth Defects Prevalence — U.S. National Estimates, 2018–2022. Centers for Disease Control and
1214 Prevention (CDC), Available at https://www.cdc.gov/birth-defects/media/pdfs/2026/04/Data-Brief-
1215 Birth-Defects-National-Estimates-2018-2022.pdf
1216 16. ICD-11 for Mortality and Morbidity Statistics. WHO, Available at https://icd.who.int/browse/2024-
1217 01/mms/en#1516623224
1218 17. ICH S6 (R1) Guideline on Preclinical safety evaluation of biotechnology-derived pharmaceuticals
1219 (EMA/CHMP/ICH/731268/1998 (ICH S6)
1220 18. EMA Reflection paper on non-human primates in safety testing of human medicinal products and
1221 opportunities for 3Rs implementation (EMA/CHMP/55697/2025)
1222 19. Guideline on good pharmacovigilance practices (GVP) Module V – Risk management systems (Rev
1223 2) (EMA/838713/2011)
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 32/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 33 页
1224 20. Rasmussen S, Jamieson D, Honein M, Petersen L. Zika Virus and Birth Defects — Reviewing the
1225 Evidence for Causality. N Engl J Med 2016;374:1981-1987
1226 21. Shepard TH. “Proof” of human teratogenicity. Teratology 1994; 50: 97-8.
1227 22. Frank C, Faber M, Stark K. Causal or not: applying the Bradford Hill aspects of evidence to the
1228 association between Zika virus and microcephaly. EMBO Mol Med 2016 March 1
1229 23. Recommendations related to contraception and pregnancy testing in clinical trials, CTCG Clinical
1230 Trials Coordination Group
1231 24. Guideline on good pharmacovigilance practices (GVP) Module XVI Addendum I – Risk minimisation
1232 measures for medicinal products with embryo-fetal risks (EMA/608947/2021)
1233 25. Hale's Medications & Mothers' Milk 2025-2026: A Manual of Lactational Pharmacology 21st ed.
1234 Edition by Thomas W Hale Rph PhD (Author), Kaytlin Krutsch PhD Pharmd MBA Bcps (Author)
1235
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product
information Page 33/46
EMA/CHMP/PRAC/203927/2005 Rev.1
第 34 页
1236 Annex
1237 Standard statements for section 4.6 of Summary of Product Characteristics (SmPC) and section 2 of Package Leaflet (PL):
1238 There are three tables with statements for the SmPC and the PL:
1239 • Table 1 - Pregnancy and Females of reproductive potential
1240 • Table 2 – Breastfeeding
1241 • Table 3 - Fertility.
1242 Tables 1 and 2 are divided into sections for each recommendation of use, ranging from a strict contraindication to that the medicinal product can be used.
1243 For each recommendation in table 1 and 2, there are different statements to be used in describing available data and risk minimisation measures, if
1244 applicable, or other relevant information for the medicinal product, based on the specific assessment (see sections 7 and 8). For fertility, standard
1245 statements to describe data are provided, since in most cases, specific recommendations are not provided in the product information in relation to fertility.
1246 Recommendations may be needed on a case-by-case basis, specifically in situations of identified risk of temporarily impaired fertility, difficulties conceiving
1247 or risk of irreversible damage in fertility (see table 3 and section 7.2.1).
1248 The structure of the product information should be in accordance with the QRD template, i.e. pregnancy, breastfeeding and fertility, with additional sub-
1249 headings in the SmPC and PL, such as “Women of childbearing potential”, “Contraception in males” or “Contraception in females”, as appropriate.
1250 Furthermore, in the QRD template, each subsection starts with description of data, followed by a recommendation and risk minimisation measures, if
1251 applicable.
1252 How to compile the standard statements
1253 SmPC
1254 The assessment of available data for a specific medicinal product, will result in a Recommendation (Section 7.2). This recommendation will be the starting
1255 point for compiling a complete wording for each subpart of section 4.6 of the SmPC. In general, standard wording should be selected from one of the
1256 recommendations only, although it can be adapted if there are no suitable wording for a specific case.
1257 To compile the standard statement, begin with the text in the recommendation-box for the applicable recommendation, and continue adding text following
1258 the steps under the appropriate recommendation (e.g. female of reproductive potential, data description, risk minimisation measures) and include the
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product information
EMA/CHMP/PRAC/203927/2005 Rev.1 Page 34/46
第 35 页
1259 standard statement that is suitable for the intended medicinal product. If a subsection is irrelevant for the intended medicinal product (e.g. risk minimisation
1260 measures), no standard statement should be included from that specific subsection.
1261 Package leaflet
1262 When compiling the information for the section 2 of the Package leaflet, use the corresponding text in relation to the standard statements used for the
1263 SmPC. In some cases, there are more options in the PL compared to the SmPC standard statements. In such case, use one or several of the PL standard
1264 statements, considering the relevance of the intended product. It is also acceptable to change the order of the selected standard statements, compared to
1265 the order in this Annex, to get a readable text without repetition of information.
1266 Note: {active substance}: there may be cases for which it is most appropriate to use {invented name} i.e.: when the risk depends on the route of
1267 administration or the formulation; N/A – Not applicable. Information is not relevant for the individual (e.g. lack or uncertainty of clinical data, non-clinical
1268 data). No statement should be included in the package leaflet.
1269 Table 1. Pregnancy and females of reproductive potential
Steps
drafting Template sentences for SmPC Template sentences for PL section 2
sentences
Recommendation 1: The medicinal product is contraindicated during pregnancy
[Section 4.3] Pregnancy <{trimester} of pregnancy> (see section [Pregnancy] Do not <take><use> this medicine <if you are
4.6) pregnant><during the <first><last> <3 months><6 months> of
[Section 4.6] Treatment with {Active substance} is pregnancy> and <up to {number} weeks <after><before>
contraindicated during <pregnancy><the {number} trimester of pregnancy> because it can harm the child.
pregnancy> and<up to {number} weeks <after><before>
pregnancy> (see section 4.3).
Step 1. Select recommendation for female of reproductive potential (select one of two options as appropriate to the product and its
indication)
Option 1 [Section 4.3] Females of reproductive potential, unless they are [Pregnancy] Do not <take> <use> this medicine if you can become
using effective contraception. pregnant, unless you are using effective birth control.
[Section 4.6 <section 4.4>] Females of reproductive potential [Pregnancy] If you can become pregnant, you must use effective birth
must use effective contraception during treatment with {active control during treatment with this medicine <and for <at least>
substance} and < for {number <days><weeks><months>}> {number < days >< weeks><months>} after the last dose of this
after treatment>. medicine>. You must not become pregnant while <taking><using>
this medicine.
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product information
EMA/CHMP/PRAC/203927/2005 Rev.1 Page 35/46
第 36 页
Steps
drafting Template sentences for SmPC Template sentences for PL section 2
sentences
[Section 4.6 <section 4.4>] Individual circumstances should be [Pregnancy] Talk to your doctor about birth control methods that may
evaluated together with the patient when choosing the be right for you. Your doctor will give you information on preventing
contraceptive method. pregnancy and may refer you to a specialist for advice on birth control.
There may be individual circumstances where a female is not
required to use effective contraception during <and after>
treatment with {active substance} following discussion with their
prescriber. This should be regularly evaluated to determine
whether circumstances change.
Option 2 [Section 4.6] In females of reproductive potential, effective [Pregnancy] If you can become pregnant, effective birth control during
contraception during treatment with {active substance} <and < treatment with this medicine <and for <at least> {<number> < days
for {<number> < days >< weeks><months>}> after treatment> >< weeks><months>} after the last dose of this medicine> is
is recommended.> recommended.
[Section 4.6] Individual circumstances should be evaluated [Pregnancy] Talk to your doctor about birth control methods that may
together with the patient when choosing the contraceptive be right for you. Your doctor may refer you to a specialist for advice on
method. birth control.
Step 2. Select most suitable text describing available data relevant for pregnancy
Clinical data [Section 4.6] Data from {type of study}, including approximately [Pregnancy] This medicine may <in <very> rare cases> increase the
{number} pregnancies, exposed during {Y trimester}, <show > risk of <birth defects <such as {list the side effects}><{main adverse
<suggest > an approximately {number}-fold increase in {main pregnancy outcome(s)}> when taken during pregnancy.
Option 1 adverse pregnancy outcome(s)} corresponding to {concrete
[Pregnancy] <In pregnant individuals who take {active substance},
data/numbers; excess risk estimates (if possible); pattern of
around {number or number interval} children in every {number} will
{main adverse pregnancy outcome(s)} observed; existence of
have birth defects. This compares with {number or number interval}
dose-related effects}, compared with {reference/comparator
children in every {number} born to individuals who do not have
group, confidence intervals and background frequency, as
{condition} and do not take a {medicine class} treatment.>
applicable}.
Option 2 [Section 4.6] Based on <human experience {describe type of
experience shortly}> <and> <or> <mechanism of action>,
{active substance} <increases> <may increase> the risk of
embryo-fetal harm when administered during pregnancy.
Option 3 [Section 4.6] There are no or limited data from the use of {active N/A
substance} in pregnant individuals.
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product information
EMA/CHMP/PRAC/203927/2005 Rev.1 Page 36/46
第 37 页
Steps
drafting Template sentences for SmPC Template sentences for PL section 2
sentences
Non-clinical [Section 4.6] Non-clinical data have shown developmental toxicity, N/A
data including <malformations><embryo-fetal
lethality><peri/postnatal mortality> (see section 5.3).
Option 1
Option 2 [Section 4.6] Based on non-clinical data <from developmental
toxicity studies and/or mechanism of action> (see section 5.3)
{active substance} <increases> <may increase> the risk for
embryo-fetal harm when administered during pregnancy.
Option 3 [Section 4.6] Non-clinical data are insufficient with respect to
developmental toxicity (see section 5.3).
Step 3. Select further risk minimisation measures (<> bracket indicates to select as appropriate to the product, risk and indication)
The alternatives below concern cases where the product is contraindicated in pregnancy and in females of reproductive potential unless they are using
effective contraception, or where the product is contraindicated in pregnancy but not in females of reproductive potential
[Pregnancy] <Read the <select additional risk minimisation material e.g.
[Section 4.6] <Educational materials for risk minimisation
measures <select relevant material e.g. checklist, guide> are patient guide, patient card> <that you will receive from your
available for <health care professional> <patients > <and <doctor><pharmacist> <find in the package> >
caregivers>. The patient <select additional risk minimisation
material e.g. patient guide or patient card> must be provided to
<select target group e.g. females of reproductive potential >.
<Pregnancy [Section 4.6 <section 4.4>] In females of reproductive potential, [Pregnancy] If you can become pregnant, you must take a pregnancy
testing> a pregnancy test must be negative before the start of treatment test that must be negative before starting treatment with this
with {active substance}. medicine.
<Counselling [Section 4.6 <section 4.4>] Females of reproductive potential [Pregnancy] Talk to your doctor about the risks of <taking><using>
> must be advised about the risks of using {active substance} this medicine during pregnancy for advice on <appropriate actions to
during pregnancy and about the risk minimisation measures to avoid becoming pregnant><and><or><birth control methods >, if you
avoid use during pregnancy, including contraceptive measures. can become pregnant.
<When [Section 4.6 <section 4.4>] If a female using {active substance} [Pregnancy] If you are planning to become pregnant during the use of
pregnancy is plans to become pregnant, treatment must be discontinued <, this medicine, treatment with this medicine must be stopped<, and, if
considered and, if needed, switched to an alternative treatment> before needed, switch to another medicine> before stopping birth control.
during contraception is discontinued. To facilitate this, <Your doctor will discuss with you your treatment needs and suitable
treatment> treatment alternatives.>Do not stop <taking><using> your birth
control before you have discussed it with your doctor.
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product information
EMA/CHMP/PRAC/203927/2005 Rev.1 Page 37/46
第 38 页
Steps
drafting Template sentences for SmPC Template sentences for PL section 2
sentences
-<the female should be advised to consult her physician as soon [Pregnancy] <Talk to your doctor <or to a specialist experienced in
as she is planning pregnancy>. {the indication for the product}> as soon as you plan to become
pregnant during treatment with this medicine.>
-<a specialist experienced in the management of {the indication
for the product} must be consulted>.
<Unplanned [Section 4.6 <section 4.4>] If an individual becomes pregnant [Pregnancy] If you discover that you are pregnant while
pregnancy while using {active substance}: <taking><using> this medicine,
during
-<the individual should be advised to urgently consult the - <talk to your doctor immediately for advice. Do not delay seeking
treatment>
physician>. medical guidance.>
-<the individual should be advised to immediately cease - <do not stop <taking><using> this medicine on your own, as your
treatment with {active substance}.> condition may become worse.>
-<the individual must be immediately referred to a specialist to - <your doctor can tell you to stop treatment immediately>
re-evaluate treatment with {active substance} and consider
alternative options>. - <your doctor will urgently refer you to a specialist experienced in
treating {the indication for the product}. The specialist will reassess
The individual should be evaluated and counselled regarding the your treatment and consider another medicine for you.>
exposed pregnancy.
- <you may also get advice about the possible effects of this medicine
on your pregnancy and your treatment options.>
Recommendation 2: Medicinal product should not be used during pregnancy unless medical condition requires treatment and cannot be
managed by an alternative treatment.
[Pregnancy] You should not <take><use> this medicine <if you are
[Section 4.6] {Active substance} should not be used during
pregnant><during the {number} trimester of pregnancy><up to
<pregnancy>< the {number} trimester><up to {number} weeks
{number} weeks <after><before> treatment> unless considered
<after><before> treatment> unless the medical condition
necessary for treating your condition, and any other treatment is not
requires treatment with {active substance} and cannot be
suitable.
managed by an alternative treatment.
Step 1. Select recommendation for female of reproductive potential (select one of two options as appropriate to the product and its
indication)
Option 1 [Section 4.6] In females of reproductive potential, effective [Pregnancy] If you can become pregnant, effective birth control during
contraception should be used during treatment with {active treatment with this medicine <and for <at least> {<number> <days
substance} and <for {number} < days >< >< weeks><months>} after the last dose of this medicine> is
weeks><months>><after treatment>> recommended.
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product information
EMA/CHMP/PRAC/203927/2005 Rev.1 Page 38/46
第 39 页
Steps
drafting Template sentences for SmPC Template sentences for PL section 2
sentences
[Section 4.6] Individual circumstances should be evaluated [Pregnancy] Talk to your doctor about birth control methods that may
together with the patient when choosing the contraceptive be right for you. Your doctor will give you information on preventing
method. pregnancy and may refer you to a specialist for advice on birth control.
Option 2 [Section 4.6] No recommendation for females of reproductive N/A
potential is warranted
Step 2. Select most suitable text describing available data
Clinical data [Section 4.6] Data from {type of study}, including approximately [Pregnancy] <In <very> rare cases>, this medicine may increase the
{number} pregnancies, exposed during {Y trimester}, <show > risk of <birth defects <such as {list the side effects}>>< {main
<suggest > an approximately {number}-fold increase in {main adverse pregnancy outcome(s)}> when taken during pregnancy.
Option 1 adverse pregnancy outcome(s)} corresponding to {concrete
data/numbers; excess risk estimates (if possible); pattern of
{main adverse pregnancy outcome(s)} observed; existence of
dose-related effects}, compared with {reference/comparator
group, confidence intervals and background frequency, as
applicable}.
Option 2 [Section 4.6] Based on <human experience {describe type of
experience shortly}> <and> <or> <mechanism of action>,
{active substance} <increases> <may increase> the risk of
embryo-fetal harm when administered during pregnancy.
Option 3 [Section 4.6] There are no or limited data from the use of {active N/A
substance} in pregnant individuals.
Non-clinical [Section 4.6] Non-clinical data have sown developmental toxicity, N/A
data including <malformations><embryo-fetal
lethality><peri/postnatal mortality> (see section 5.3).
Option 1
Option 2 [Section 4.6] Based on non-clinical data <from developmental
toxicity studies and/or mechanism of action> (see section 5.3)
{active substance} <increases> <may increase> the risk for
embryo-fetal harm when administered during pregnancy.
Option 3 [Section 4.6] Non-clinical data are insufficient with respect to
developmental toxicity (see section 5.3).
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product information
EMA/CHMP/PRAC/203927/2005 Rev.1 Page 39/46
第 40 页
Steps
drafting Template sentences for SmPC Template sentences for PL section 2
sentences
Option 4 [Section 4.6] Non-clinical data do not indicate developmental
toxicity (see section 5.3).
Step 3. Select further risk minimisation measures (<> bracket indicates to select as appropriate to the product, risk and indication)
<Pregnancy [Section 4.6 <section 4.4>] <In females of reproductive potential, [Pregnancy] <If you can become pregnant, you should considering take
testing> a pregnancy test should be considered before starting treatment.> a pregnancy test before starting treatment with this medicine.>
<Counselling [Section 4.6 <section 4.4>] Females of reproductive potential [Pregnancy] Your doctor should tell you about risks of
> must be advised about the risks of using {active substance} <taking><using> this medicine during pregnancy and should advice
during pregnancy and about the risk minimisation measures to you about use during pregnancy, if you can become pregnant.
avoid use during pregnancy <including contraceptive measures>.
<When [Section 4.6 <section 4.4>] If an individual using {active [Pregnancy] If you are planning to become pregnant during the use of
pregnancy is substance} plans to become pregnant, treatment may need to be this medicine, treatment with this medicine may need to be stopped <,
considered discontinued <, and, if needed, switched to an alternative and, if needed, switched to another medicine> before stopping birth
during treatment> before contraception is discontinued. To facilitate this, control. < Your doctor will discuss with you your treatment needs and
treatment> suitable treatment alternatives.> Do not stop <taking><using> your
-<the individual should be advised to consult the physician as
birth control method before you have discussed it with your doctor.
soon as planning pregnancy>.
[Pregnancy] <Talk to your doctor <or to a specialist experienced in
-<a specialist experienced in the management of {the indication
{the indication for the product}> as soon as you plan to become
for the product} may be consulted>.
pregnant.>
[Pregnancy] <Discuss <with your doctor> whether <taking><using>
this medicine during pregnancy is suitable for you, considering your
treatment needs and any potential side effects to the child. The doctor
may suggest switching to another medicine.>
<Unplanned [Section 4.6 <section 4.4>] If an individual becomes pregnant [Pregnancy] If you discover that you are pregnant while taking this
pregnancy while using {active substance}: medicine,
during
-<the individual should be advised to urgently consult the - <talk to your doctor immediately for advice. Do not delay seeking
treatment>
physician>. medical guidance.>
-<the individual should be advised to immediately cease - <do not stop <taking><using> this medicine on you own, as your
treatment with {active substance}.> condition may become worse. >
- <your doctor may tell you to stop treatment immediately>
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product information
EMA/CHMP/PRAC/203927/2005 Rev.1 Page 40/46
第 41 页
Steps
drafting Template sentences for SmPC Template sentences for PL section 2
sentences
-<the individual must be immediately referred to a specialist to -<your doctor will urgently refer you to a specialist experienced in
re-evaluate treatment with {active substance} and consider treating {the indication for the product}. The specialist will reassess
alternative options>. your treatment and consider another medicine for you.
The individual should be referred for evaluation and counselling [Pregnancy] < Discuss <with your doctor> whether <taking><using>
regarding the exposed pregnancy. this medicine during pregnancy is suitable for you, considering your
treatment needs and any potential side effects to the child. The doctor
may suggest switching to another medicine.>
Recommendation 3: Use of the medicinal product during pregnancy may be considered
[Pregnancy] Discuss with your doctor whether this medicine is right for
[Section 4.6] The use of {active substance} may be considered
<during pregnancy><during the {number} trimester of you during pregnancy, considering your treatment needs and any
pregnancy>. potential side effects to the child
Step 1. Select most suitable text describing available data
Clinical data [Section 4.6] Data from {type of study}, including approximately [Pregnancy] The effects of this medicine on a child are yet not clear,
{number} pregnancies, exposed during {number} trimester, do but current knowledge suggests that there is no risk to the child.
Option 1
not suggest an increased risk of adverse pregnancy outcomes;
[Pregnancy] Use of this medicine may therefore be considered <during
however, available data are insufficient to draw firm conclusions.
pregnancy><during the <first><last> <3 months><6 months> of
pregnancy><up to {number} weeks <after><before> treatment>.
Option 2 [Section 4.6] There are no or limited data from the use of {active N/A
substance} in pregnant individuals.
Non-clinical [Section 4.6] Non-clinical data are insufficient with respect to N/A
data developmental toxicity (see section 5.3).
Option 1
Option 2 [Section 4.6] Non-clinical data have shown developmental toxicity
not considered clinically relevant (see section 5.3).
Option 3 [Section 4.6] Non-clinical data do not indicate developmental
toxicity (see section 5.3).
Recommendation 4: Medicinal product can be used during pregnancy
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product information
EMA/CHMP/PRAC/203927/2005 Rev.1 Page 41/46
第 42 页
Steps
drafting Template sentences for SmPC Template sentences for PL section 2
sentences
[Pregnancy] You can use this medicine during <pregnancy> <the
[Section 4.6] {Active substance} can be used <during
pregnancy><during {trimester} of pregnancy > <first><last><3 months><6 months> of pregnancy>.
Step 1. Select most suitable text describing available data
Clinical data [Section 4.6] Data from {type of study}, including approximately N/A
{number} pregnancies, exposed during {number trimester}, do
Option 1
not indicate an increased risk of {main adverse pregnancy
outcome(s)}, based on {brief summary of findings}.
Option 2 [Section 4.6] A large amount of data from pregnant individuals
{approximate number} indicates no increased risk of adverse
embryo-fetal or neonatal effects.
Option 3 [Section 4.6] No effects are anticipated during pregnancy since
systemic exposure to {active substance} is negligible.
1270 Table 2. Breastfeeding
Steps
drafting Template sentences for SmPC Template sentences for PL section 2
sentences
Recommendation 1: The medicinal product is contraindicated during breastfeeding
[Section 4.3] Breastfeeding (see section 4.6). [Do not <take> <use> X] If you are breastfeeding.
[Breastfeeding] Do not breastfeed during treatment with this medicine.
[Section 4.6] {Active substance} is contraindicated during
breastfeeding (see section 4.3). Breastfeeding must be You must stop breastfeeding before treatment with this medicine.
discontinued during treatment with {active substance}.
Step 1. Select most suitable text describing available data
[Section 4.6] <{Active substance} <metabolites> have been [Breastfeeding] <This medicine passes into breast milk in high
Clinical data
detected in human breast milk in high amount. Serious amounts that reach the breastfed child. Serious side effects <such as
Option 1 adverse reactions <such as {give examples of reported {give examples of reported side effects}> have been reported in
adverse reactions}> have been reported in breastfed infants of breastfed children whose mothers used <invented name>.>
treated mothers.> <Although this medicine passes into breast milk in low amounts,
<Although {Active substance} <metabolites> have been serious side effects <such as {give examples of reported side
detected in human breast milk in low amount, serious adverse
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product information
EMA/CHMP/PRAC/203927/2005 Rev.1 Page 42/46
第 43 页
Steps
drafting Template sentences for SmPC Template sentences for PL section 2
sentences
reactions <such as {give examples of reported adverse effects}> have been reported in breastfed children whose mothers
reactions}> have been reported in breastfed infants of treated used <invented name>.>
mothers.> [Breastfeeding] <This medicine passes into breast milk and is likely
<{Active substance} <metabolites> are excreted into human to cause serious side effects <such as {give examples of reported
milk to such an extent that serious adverse reactions in the side effects}> in a breastfed child.>
breastfed infant are likely.>
[Section 4.6] <Physicochemical data suggest excretion of
Option 2
{active substance} <metabolites> in breast milk><which
could lead to clinically relevant systemic exposure of the
breastfed infant >.>
<Serious adverse reactions in the breastfed infant are likely>.
Recommendation 2: Medicinal product should not be used during breastfeeding
[Section 4.6] {Active substance} should not be used during [Breastfeeding] <This medicine should not be used during
breastfeeding.[OR] <Breastfeeding should be discontinued breastfeeding.> <You should stop breastfeeding before treatment with
during treatment with {active substance}> this medicine.>
[Breastfeeding] Do not breastfeed during treatment with this medicine.
You must stop breastfeeding before treatment with this medicine.
Step 1. Select most suitable text describing available data
[Section 4.6] <{Active substance} <metabolites> have been [Breastfeeding] <This medicine passes into breast milk
Clinical data
detected in human breast milk in <minimal> <low> <high> <minimal><low><high> amounts and reaches the breastfed child.
Option 1 amount. Adverse reactions <such as {give examples of Side effects <such give examples of reported side effects}> have
reported adverse reactions}> have been reported in breastfed been reported in breastfed children whose mothers used this
infants of treated mothers.> medicine.>
<{Active substance} <metabolites> are excreted into human <This medicine passes into breast milk and is possible to cause side
milk to such an extent that adverse reactions on the breastfed effects <such as {give examples of reported side effects}> in the
infant are possible.> breastfed child.>
[Section 4.6] <Physicochemical data suggest excretion of
Option 2
{active substance} <metabolites> in breast milk><which
could lead to clinically relevant systemic exposure of the
breastfed infant.>
<Adverse reactions in the breastfed infant are possible.>
[Section 4.6] It is unknown whether {active substance} N/A
Non-clinical
<metabolites> are excreted in human milk. Non-clinical data
data
have shown toxicity <from developmental toxicity studies
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product information
EMA/CHMP/PRAC/203927/2005 Rev.1 Page 43/46
第 44 页
Steps
drafting Template sentences for SmPC Template sentences for PL section 2
sentences
and/or mechanism of action><such as {concrete adverse
Option 1
effects}> (see section 5.3).
Recommendation 3: Use of the medicinal product during breastfeeding may be considered
[Section 4.6] The use of {active substance} may be considered [Breastfeeding] Discuss with your doctor whether this medicine is right
during breastfeeding. A decision whether to breastfeed while for you while breastfeeding, considering the benefits of breastfeeding
using {active substance} should take into account both the and any potential side effects for your child.
benefit of breastfeeding and the potential risk of adverse
reactions in the infant.
Step 1. Select most suitable text describing available data
[Section 4.6] <It is unknown whether {active substance} [Breastfeeding] <It is unknown if this medicine passes into the
Clinical data
<metabolites> are excreted in human milk.> breast milk.>
Option 1 Alternatively: < {active substance} <metabolites> are Alternatively: <This medicine passes into the breast milk in low
excreted in human milk in low levels.> amounts.>
<Serious adverse reactions on a breastfed infant are not <Serious side effects in the breastfed child are not expected.>
expected>
[Section 4.6] <{Active substance} is not expected to be [Breastfeeding] < This medicine passes into breast milk, however, it
Option 2
absorbed from infant´s gastrointestinal tract.> is not expected to pass from your child's stomach and intestines into
their blood stream.>
[Section 4.6] <Physicochemical data suggest excretion of [Breastfeeding] <This medicine may pass into the breast milk.>
Option 3
{active substance} <metabolites> in breast milk><which <Serious side effects in the breastfed child are not expected.>
could lead to clinically relevant systemic exposure of the Breastfeeding may be considered while you are <taking><using>
breastfed infant> this medicine.
<Serious adverse reactions on a breastfed infant are not
expected>.
Recommendation 4: Medicinal product can be used during breastfeeding
[Section 4.6] {Active substance} can be used during [Breastfeeding] You can use this medicine while breastfeeding.
breastfeeding.
Step 1. Select most suitable text describing available data
[Section 4.6] <No effects of {active substance} have been [Breastfeeding] <No side effects have been shown in breastfed
Clinical data
shown in breastfed infants of treated mothers.> children whose mothers <used><received> this medicine.>
Option 1
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product information
EMA/CHMP/PRAC/203927/2005 Rev.1 Page 44/46
第 45 页
Steps
drafting Template sentences for SmPC Template sentences for PL section 2
sentences
[Section 4.6] <No effects on the breastfed infant are [Breastfeeding] <Side effects in the breastfed child are not
Option 2
anticipated since the systemic exposure of {active substance} anticipated.>
on the breastfed infant is negligible.>
[Section 4.6] <{Active substance} <metabolites> are [Breastfeeding] <Serious side effects in the breastfed child are
Option 3
excreted in human milk, but at therapeutic doses of {active unlikely.>
substance} no effects on the breastfed infant are anticipated.>
<Serious adverse reactions on the breastfed infant are
unlikely.>
Breastfeeding recommendations for specific medicinal products
Monoclonal [Section 4.6] <It is unknown whether {active substance} is [Breastfeeding] <It is not known whether this medicine passes into
antibodies excreted in human breast milk. Human IgGs are known to be breast milk. Some natural antibodies (a type of protein) pass into
excreted in breast milk for a few days after birth, declining to breast milk for a few days after birth, but their levels drop quickly
low levels thereafter. Therefore, exposure of the breastfed after this time. Because of this, a possible risk to a breastfeeding
infant <and a risk for adverse reactions> is possible during the newborn during those first few days after birth cannot be fully ruled
initial days after birth.> out. >
[Considering the major elements of assessment, the following
sentence may also be included] [breastfeeding] <After the first few days, breastfeeding may be
[Breastfeeding] <Beyond this period, use of {active considered>
substance} during breastfeeding may be considered.>
[Breastfeeding] <Talk to your doctor about whether to use this
medicine while breastfeeding, considering the benefits of
breastfeeding, your treatment needs, and the potential risks of side
effects to your child.>
Protein- [Section 4.6] <It is unknown whether {active substance} is [Breastfeeding] <It is not known whether this medicine passes into
based excreted in human breast milk. However, protein-based breast milk. However, medicines made from proteins are expected to
therapeutics products are expected to be denatured and degraded in the be broken down in the child’s body and are unlikely to affect the
infant’s gastrointestinal tract.> child.
[Considering the major elements of assessment, the following Because of this, breastfeeding can be considered while you are
sentence may also be included] <taking><using> this medicine.>
[Breastfeeding] < Use of {active substance} during
<Talk to r your doctor to discuss if you should use this medicine
breastfeeding may be considered.>
while breastfeeding, considering the benefits of breastfeeding, your
treatment needs, and the potential risks of side effects to your
child.>
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product information
EMA/CHMP/PRAC/203927/2005 Rev.1 Page 45/46
第 46 页
1271 Table 3. Fertility
Steps
drafting Template sentences for SmPC Template sentences for PL section 2
sentences
Step 1. Select most suitable text describing available data
[Section 4.6] <Based on the mechanism of action of {active There is a risk that this medicine <temporarily><permanently>
Clinical data
substance}, there is a risk for impaired fertility in affects your ability to <get pregnant><father a child>.
Option 1 <both><males> <and> <or> <females>.>
<Discuss with your doctor before you <take><use> this medicine if
[Section 4.6] <Clinical data <show> <suggest> risk of you are planning to have a child.>
Option 2
<reversible><irreversible> <impaired fertility> after
treatment with {active substance} in <both><males> <and>
<or> <females>.
[Section 4.6] <There is no human data with respect to
Option 3 No statement needed
fertility.>
[Section 4.6] <Non-clinical data with {active substance} have There is a risk that this medicine <temporarily><permanently>
Non-clinical
shown <reversible> <irreversible> adverse effects on fertility affects your ability to <get pregnant><father a child>.
data
in <both><males> <and> <or> <females> (see section
<Discuss with your doctor before you <take><use> this medicine if
Option 1 5.3).>
you are planning to have a child.>
[Section 4.6] <Findings from non-clinical data indicate that
Option 2 N/A
{active substance} may impair <both><males> <and> <or>
<females> fertility (see section 5.3).>
[Section 4.6] <Fertility studies with {active substance} were
Option 3
not performed in animals.>
[Section 4.6] < Findings from non-clinical data with {active
Option 4
substance} are inconclusive on <both><males> <and> <or>
<females> fertility (see section 5.3).>
[Section 4.6] <Non-clinical data with {active substance}
Option 5
<have not shown><do not suggest> any adverse effects on
fertility in <both><males> <and> <or> <females>> (see
section 5.3).
1272
Guideline on risk assessment of medicinal products in human reproduction and breastfeeding: from data to product information
EMA/CHMP/PRAC/203927/2005 Rev.1 Page 46/46
来源:EMA Scientific Guidelines · ema.europa.eu