Specific effects on the reproductive function including fertility relevant for cumulative risk assessment of pesticide residues Authors: European Food Safety Authority (EFSA), Karine Angeli, Adeline Cavelier, Tamara Coja, Federica Crivellente, Anna Lanzoni, Luc Mohimont, Monica Nepal, Dimitra Nikolopoulou, Andrea Terron, Anna F. Castoldi Published: 2025 ABSTRACT AbstractAccording to the ‘EFSA‐SANTE Action Plan on Cumulative Risk Assessment for pesticides residues’, EFSA, with the support of a working group, undertook the identification of specific effects on male and female reproductive function, including fertility, that are considered relevant for grouping pesticide residues and for assessing retrospectively the cumulative risk (CRA) deriving from consumers' dietary exposure. In this first report, two specific effects leading to two correspondent cumulative assessment groups (CAGs) were identified for male reproduction, namely, (1) altered mating performance and (2) decreased fertility. Four distinct CAGs were proposed for grouping chemicals affecting female reproduction: (1) altered mating performance, (2) decreased fertility, (3) altered gestation and (4) altered parturition. EFSA also defined the list of indicators (i.e. toxicological endpoints measurable in regulatory studies) describing each specific effect. Qualitative and quantitative information on these indicators will be collected from the toxicological assessment reports on individual pesticide active substances: indicators will be used to determine the inclusion of prioritised active substances and their metabolites into the respective CAG and to estimate the no observed adverse effect level (NOAEL) and lowest observed adverse effect level (LOAEL) of each chemical in relation to the given specific effect. Additional endpoints (named ancillary findings), not sufficiently informative to define a specific effect but contributing to the overall evidence, will be collected for a limited number of substances identified as risk drivers, based on hazard and exposure considerations, to support the estimation of CAG‐membership probabilities. Additional criteria for including active substances into CAGs, the hazard characterisation methodology and the lines of evidence for assessing CAG‐membership probabilities are detailed in this report, which was sent for public consultation. The process of data extraction and actual establishment of the CAGs are instead beyond the scope of this report. This part of the CRA process was outsourced and will be dealt with in a separate report. SUMMARY EFSA was requested to perform a retrospective dietary cumulative risk assessment (CRA) for pesticide residues targeting the reproductive function including fertility, according to the agreed 2021 ‘EFSA‐SANTE Action Plan on Cumulative Risk Assessment for pesticides residues’. Due to the complexity of the CRA process, the full cycle of this CRA, which covers the four pillars of risk assessment (hazard identification, hazard characterisation, exposure assessment and risk characterisation), will require several sequential steps. The outcome of the whole work will therefore be presented in four separate reports published at different times along the CRA process. The present report developed by an EFSA Working Group (WG) focuses on the first CRA step, i.e. hazard identification around reproductive function including fertility. It aims at defining the specific effects (i.e. precise toxicological insults) on male and female reproductive functions including fertility, that are relevant for grouping prioritised pesticide active substances (a.s.) and/or metabolites into cumulative assessment groups (CAGs) and for assessing retrospectively the cumulative risks deriving from consumers' dietary exposure. This document also defines the indicators of each specific effect (i.e. toxicological endpoints measurable in regulatory studies) used for the allocation of a.s. and/or metabolites to the respective CAG. It also describes in detail the methodology for setting up the CAGs, for characterising the pesticide residues included in these CAGs and for assigning to each of them a CAG membership probability. Implementation of this methodology and actual establishment of the CAGs, from the data extraction onwards, are beyond the scope of this document. According to the EFSA Scientific Committee's framework, the golden standard for grouping chemicals is a common mode of action (MoA) and/or adverse outcome pathway (AOP). Overall, 32 relevant AOPs were identified as relevant for reproductive specific effects from the AOPwiki (Version 2.7, released on 30 March 2024): of these, 21 are applicable to males, 10 to females and one, the AOP492, to both sexes. Unfortunately, none has been endorsed by the OECD except for AOP212 (Histone deacetylase inhibition leading to testicular atrophy). Although an endocrine‐disrupting (ED) MoA has been postulated for substances for which the ED assessment is performed according to ECHA/EFSA Guidance, since this information (ED MoA or not) is not available for all substances, grouping based on ED MoA is not possible. Considering the limitations of the relevant AOPs mentioned above and the lack of information on the MoA leading to effects on reproductive function and fertility for several substances, the selected strategy was to group a.s. and metabolites into CAGs based on common adverse outcomes (AOs) captured as apical toxicity endpoints in regulatory studies and referred to as specific effects in the present report. For the male reproductive function, two specific effects leading to two correspondent CAGs were proposed, namely (1) altered mating performance and (2) decreased fertility. Four CAGs were suggested for grouping chemicals affecting the female reproductive function, i.e. (1) altered mating performance, (2) decreased fertility, (3) altered gestation and (4) altered parturition. A list of indicators describing each specific effect was compiled separately for males and females. Examples of indicators are e.g. ‘Delayed time to mating’ and ‘decrease in male/female mating index’ for the allocation of a substance to the two CAGs for the specific effect ‘altered mating performance’ in males and females, respectively. Furthermore, for the CAGs, ‘decreased fertility’ different indicators were selected for males and females. For example, in males, these indicators were decreased male fertility index, decreased number of implantation sites, decreased litter size, changes in sperm parameters and changes in testes histopathology. The full list of indicators identified for each specific effect on male and female reproductive functions is presented in a dedicated section of the report. The methodology described in this report foresees that relevant qualitative and quantitative information and data on prioritised individual substances will have to be extracted from the toxicity studies included in the draft assessment reports (DARs), renewal assessment reports (RARs) and/or other similar regulatory documents. The list of prioritised individual substances has been developed from the list of pesticide a.s. quantified in at least one food sample of the official monitoring programme of EU Member States over a 3‐year monitoring cycle from 2019 to 2021; the EFSA prioritisation methodology (EFSA, 2024) was applied and reproductive toxicity checked as identified in an EFSA peer review (if not yet fully concluded, at least discussed and agreed in an Experts' meeting). In addition, the list of prioritised substances is complemented by pesticide metabolites present in food and that have a reproductive toxicity profile. Data and information will be collected in an ad hoc data collection template developed by the EFSA WG in collaboration with the Austrian Agency for Health and Food Safety (AGES, Vienna, Austria) under a framework partnership agreement. The presence of one or more indicator(s) of the respective specific effect in the data collection will be used as the main criterion to trigger the inclusion of a given prioritised a.s. and/or metabolite into the respective CAG. Each a.s. or metabolite included in a CAG will be characterised by the identification of an overall no observed adverse effect level (NOAEL) and an overall lowest observed adverse effect level (LOAEL) for the respective specific effect based on all available data from relevant repeated dose oral toxicity studies. Further details on the criteria applied for including substances into CAGs and for the hazard characterisation methodology are given in the report. It is acknowledged that the amount, reliability, relevance and consistency of the evidence causally linking a substance to a specific effect vary between different substances, making the conclusion on the substances that actually belong to a CAG uncertain. This uncertainty, i.e. CAG‐membership probability, can be quantified by a weight of evidence (WoE) approach, estimating the probability that any substance actually causes the specific effect. This will be assessed for a limited number of substances per CAG, so‐called ‘risk drivers’, 1 selected on the basis of hazard and exposure considerations. The list of risk driver a.s. will be defined at the exposure assessment step and the underlying methodology will be fully described in the exposure report (also see Appendix A, Table A.1, step 13). A comprehensive list of lines of evidence (LoE), which were assigned by the WG a different level (on a 5‐degree scale) of strength, will be used to assess the CAG membership probabilities of risk drivers. For these substances, the data collection will be expanded to cover endpoints other than indicators, so‐called ‘ancillary’ findings. These refer to changes in male/female reproductive parameters (e.g. changes in reproductive organ weight) that in isolation are not sufficient to support the direct inclusion of an a.s. into a CAG. However, if they are detected in studies where at least one indicator has been observed, ancillary findings can support the indicator(s) and increase the likelihood that the substance has been correctly allocated to a CAG. Some recommendations concerning research on AOPs, investing in toxicity testing, better data reporting and data analysis are also included. This report was sent for public consultation (see Annex B). Accounting for the relevant comments received and further discussions in the WG, the specific effects of male and female reproductive function, their respective indicators and ancillary findings, the criteria for inclusion of a substance into a CAG, the hazard characterisation methodology, the lines of evidence for CAG‐membership probability and the data collection template were revised as appropriate and finalised for publication. INTRODUCTION Cumulative risk assessment (CRA) has been defined as the analysis, characterisation and possible quantification of the combined risks to health or the environment from multiple agents or stressors (U.S. EPA, 2003). It differs from most assessments which consider the effects of one agent or stressor in isolation. In Europe, Regulation (EC) No 396/2005 on maximum residue levels (MRLs) in or on food and feed of plant and animal origin 2 and Regulation (EC) No 1107/2009 concerning the placing on the market of plant protection products 3 requires thorough consideration of cumulative and potential synergistic effects of residues of plant protection products (PPPs). To comply with these provisions, in 2007, EFSA and the Panel on plant protection products and their residues (PPR Panel) started the development of the necessary methodologies to carry out CRA of pesticide residues. This methodological development included a tiered approach for the assessment of cumulative risks of pesticide residues (EFSA PPR Panel, 2008), a guidance on the use of probabilistic methodology for modelling dietary exposure to pesticide residues (EFSA PPR Panel, 2012) and a procedure to establish cumulative assessment groups (CAGs) of pesticides on the basis of their toxicological profile (EFSA PPR Panel, 2013). In the latter opinion, the PPR Panel concluded that synergistic effects are less likely to occur at low doses/concentrations corresponding to the dietary exposure to mixtures of pesticides and are thus less relevant for CRA of pesticide residues in food (EFSA PPR Panel, 2013). In addition, the limited available empirical evidence suggests that synergisms at dietary exposure levels are rather rare, and that they experimentally occurred at unrealistically high concentrations (Hernández et al., 2017). In 2020, EFSA issued the first two reports on retrospective CRA of pesticides that have acute effects on the nervous system and chronic effects on the thyroid gland (EFSA, 2020a, 2020b). These were followed in 2021 by a retrospective CRA regarding chronic acetyl cholinesterase inhibition (EFSA, 2021) and in 2022 by a retrospective CRA regarding craniofacial alterations (EFSA, 2022). Ongoing work comprises the retrospective assessment of the cumulative risks from co‐exposure to pesticide residues on the kidney, and on the liver, as well as the repetition of the retrospective CRA of pesticides affecting the thyroid based on updated data on exposure and toxicology. In 2021, EFSA and SANTE agreed on a long‐term action plan on CRA for pesticide residues (EFSA‐SANTE, 2021). This plan entails that all toxicological effects of pesticides of relevance for CRA have to be identified by 2030 and CAGs for the respective organs and systems have to be established. This plan relies on two main pillars: prioritisation of active substances (a.s.) and organs/systems and cooperation with Member States' competent organisations. Background and Terms of Reference as provided by EFSA As part of the 2021 long‐term EFSA‐SANTE action plan, EFSA decided to perform a CRA for their cumulative effects on the function of the reproductive system including fertility. This was motivated by the incidence and severity of reproductive effects observed in toxicological studies (EFSA, 2024) and the plausibility for a combined action of pesticides at this target system. The entire CRA process is planned over indicatively 27 months. It consists of a sequence of steps which are described in Figure 1. These 14 steps cover the standard scheme of risk assessment (hazard identification and characterisation, exposure assessment and risk characterisation) with periods of consultations (public consultation, PPR Panel consultation). The 14 steps of the CRA of pesticide residues. The content of each step is described in detail in Appendix A. The CRA process is performed via the cooperation with Member States, and it is documented in the following outputs: Output 1 (the present report): EFSA Scientific Report defining the specific effects on the reproductive function including fertility relevant for performing a dietary CRA of pesticide residues (referred to as ‘specific effects’ in the rest of the present report). The preparation of this report is supported by the EFSA Working Group (WG) on CRA. Output 2: Scientific Report prepared by the Austrian Agency for Health and Food Safety (AGES, Vienna, Austria), under the Framework Partnership Agreement (FPA) GP/EFSA/PREV/2021/01 on the hazard characterisation (covering the data collection) and establishment of CAGs of pesticides for specific effects on the reproductive system function including fertility. Output 3: Scientific Report of the Dutch National Institute for Public Health and the Environment (RIVM) consortium on the cumulative dietary exposure assessment of pesticides that have effects on the reproductive system function including fertility using Monte Carlo Risk Assessment (MCRA) software. Output 4: EFSA Scientific Report on the uncertainty analysis and the cumulative dietary risk characterisation of pesticides that have effects on the reproductive system function. The comments submitted on outputs 1 and 2 during the public consultations are addressed in an annex of the respective scientific reports. Purpose and content of the present report The present report (output 1) covers the identification of the specific effects of relevance for the CRA of the effects of pesticide residues on the reproductive system including fertility. Furthermore, in preparation of the establishment of CAGs, it also defines: the indicators relevant for these specific effects;the conditions (hazard identification criteria) triggering the inclusion of a.s. and metabolites into the respective CAGs;the principles for the hazard characterisation of a.s. and metabolites included in the CAGs for the respective specific effects;the lines of evidence (LoE) for the assessment of the probability that any substance included in a CAG is causing the respective specific effect (referred to as CAG‐membership probability in the rest of this report). the indicators relevant for these specific effects; the indicators relevant for these specific effects; the conditions (hazard identification criteria) triggering the inclusion of a.s. and metabolites into the respective CAGs; the conditions (hazard identification criteria) triggering the inclusion of a.s. and metabolites into the respective CAGs; the principles for the hazard characterisation of a.s. and metabolites included in the CAGs for the respective specific effects; the principles for the hazard characterisation of a.s. and metabolites included in the CAGs for the respective specific effects; the lines of evidence (LoE) for the assessment of the probability that any substance included in a CAG is causing the respective specific effect (referred to as CAG‐membership probability in the rest of this report). the lines of evidence (LoE) for the assessment of the probability that any substance included in a CAG is causing the respective specific effect (referred to as CAG‐membership probability in the rest of this report). A preliminary data collection template for CAG establishment has also been jointly developed by this WG and AGES. However, as this template will only be finalised after the public consultation of the present report, it will be made publicly available in the context of the publication of output 2 by AGES (see above). For better clarity, the scope of this report is to provide the detailed methodology for setting up the CAGs and for characterising the substances included in these CAGs. The actual establishment of the CAGs, from the data extraction onwards, has instead been outsourced to AGES (see Section 4) and will be published separately. The uncertainty analysis is also beyond the scope of this document. BACKGROUND INFORMATION FOR SETTING SPECIFIC EFFECTS ON THE REPRODUCTIVE SYSTEM INCLUDING FERTILITY RELEVANT FOR PERFORMING A DIETARY CUMULATIVE RISK ASSESSMENT OF PESTICIDE RESIDUES Reproductive system: Anatomy, physiology and function in mammals The reproductive system comprises an intricate network of organs and structures functioning collectively to ensure the continuation of the species. In mammals, the organisation, morphology and function of male and female reproductive organs exhibit differences. Male reproductive system and fertility During embryogenesis (early in the second week of gestation for the rodent), primordial germ cells (PGC or gonocytes) migrating from the yolk sac wall organise in gonadal chords. Under the guidance of SRY (sex‐determining region Y gene on the Y chromosome), these initiate to form testicular structures. Early Sertoli cells in the gonadal cords surround the centrally located gonocytes, and both multiply maximally during the second half of gestation. Leydig cells occur after cord formation, starting gestational day (GD) 15 in the rat with proliferation continuing to GD 19, followed by regression continuing to postnatal day (PND) 4. High levels of testosterone produced by the fetal Leydig cells influence reproductive organogenesis late in gestation. Postnatal male reproductive system development varies among species, as the timing of puberty varies across species, and is characterised by spermatogonia, spermatocytes and spermatids development (for the rat, see Figure 2). Development of the male reproductive system in the rat (Source: Sahota PS, Popp JA, Hardisty JF, Gopinath C and Bouchard P (Eds.). (2018). Histopathology of the uterus: Inflammation. In Toxicologic pathology: Nonclinical safety assessment (2nd ed., Section 20.14.1.4). Boca Raton, FL: Taylor & Francis). In the adult male, the overall function of the reproductive system is to produce gametes (spermatozoa) capable of fertilisation, with consequent production of viable offspring. The key organ in this process is the testis. Testis is anatomically and functionally similar across the mammalian species, including humans. In the adult, the testis is located in the extra‐abdominal scrotal sac. It is covered by the thick, fibrous tunica albuginea and the visceral tunica vaginalis. The testis contains seminiferous tubules compactly arranged in convoluted loops; in the interstitium, among tubules, there are Leydig cells, macrophages, blood vessels and nerves and supporting stroma; the testis contains a fluid (interstitial testicular fluid) rich in protein and testosterone. The arrangement of seminiferous tubules slightly differs across mammalian species: in rodents, they are arranged circumferentially, whereas in dogs and primates, they are organised in lobules. The seminiferous tubules are lined by the seminiferous epithelium composed of basally located Sertoli cells supporting maturing germ cells; these are organised in synchronised populations at different maturation stages (spermatogonia, spermatocytes, round spermatids and elongating spermatids). Seminiferous tubules converge through the tubuli recti at each end of the tubular loops in the rete testis, which is continuous with the efferent ductules and epididymis. The production of sperm through spermatogenesis and the synthesis of androgens are regulated by the hypothalamic–pituitary–gonadal (HPG) axis, which is further elaborated in the following section (OECD Guidance Document (GD) No. 106, 2009; Creasy et al., 2013). Once released from the seminiferous epithelium, spermatozoa reach the epididymis via the rete testis and ductules efferentes, where they complete maturation, acquiring motility as well as fertilising ability. The epididymis also plays crucial roles encompassing sperm transport, concentration, protection and long‐term storage in its tail (i.e. cauda epididymis) (Cornwall, 2009; Cornwall et al., 2007; Hermo & Robaire, 2002; James et al., 2020; Sullivan & Mieusset, 2016). The vas deferens serves as a conduit for sperm connecting the tail of the epididymis to the urethra. The accessory sex glands include the prostate, coagulating gland (anterior prostate), seminal vesicle, bulbourethral gland and preputial gland. There are species‐specific differences in the presence and appearance of these glands. For example, ampullary and preputial glands do not have human counterparts. The human prostate is organised into distinct zones, whereas the rodent prostate contains multiple lobes and a pair of coagulating glands (anterior lobes) adjacent to the seminal vesicles (Ittmann, 2018). Accessory glands produce secretions that are added to the spermatozoa following sexual stimulation and during ejaculation, contributing to the formation of the seminal fluid. Hormonal regulation of the male reproductive system (see Bennett et al., 2013; Clement & Giuliano, 2015; Dwyer & Quinton, 2019; OECD GD No. 106, 2009). There are various levels of hormonal regulation of the reproductive tissues, encompassing the endocrine control involving the HPG axis, as well as another tier of regulation that involves autocrine and paracrine interactions among the Sertoli cells, germ cells, Leydig cells, peritubular cells, testicular macrophages and endothelial cells of the interstitial vasculature. In the classical pathway, the cascade of the hypothalamic–pituitary–gonadal (HPG) axis involves a series of interactions among three main components: the hypothalamus, the pituitary gland and the testes. Gonadotropin‐releasing hormone (GnRH) is secreted by the hypothalamus, and it is transported through the bloodstream to the anterior pituitary gland (adenohypophysis), where it stimulates the synthesis and release of two key gonadotropins, namely the luteinising hormone (LH) and the follicle‐stimulating hormone (FSH), in a pulsatile manner. LH primarily acts on specific receptors expressed on Leydig cells to activate the production and secretion of sex hormones, particularly testosterone (the primary male sex hormone) and dihydrotestosterone (DHT). Testosterone exerts both intratesticular effects, specifically on spermatogenesis, and peripheral effects, which extend to accessory sex organs and non‐reproductive organs such as the brain, muscle, bone and skin. In many peripheral tissues, testosterone undergoes metabolism, including the conversion to oestradiol, which is a significant reaction in species other than rodents. Feedback inhibition of LH and hypothalamic GnRH is mediated by circulating levels of testosterone, DHT and oestradiol, with the relative importance of these molecules varying depending on the species. On the other hand, FSH primarily acts on the Sertoli cells within the testes, stimulating the growth and maturation of sperm cells within the seminiferous tubules. Prolonged activation of GnRH receptors can result in receptor desensitisation. In this pathway, testosterone exerts negative feedback on LH release, while inhibins and activins regulate FSH signalling and sperm production. This cascade of events within the HPG axis ensures the proper regulation of male reproductive functions, including sperm production and testosterone synthesis, which are crucial for male fertility and overall reproductive health (Figure 3). Endocrine and paracrine regulation of testicular function (Sources: OECD GD No. 106 (2009); Clement P and Giuliano F (2015). Anatomy and physiology of genital organs – men. In D.B. Vodušek and F. Boller (Eds.), Handbook of Clinical Neurology, Vol. 130 (3rd series), Neurology of Sexual and Bladder Disorders. Elsevier B.V.). This figure shows that the interplay between hormonal (endocrine) mechanisms involving GnRH, FSH, LH, testosterone and local (paracrine) signalling involves the secretion of various peptides and growth factors that regulate cellular functions through interactions among different testicular cells, including Sertoli cells (SC), germ cells (GC), peritubular myoid cells (M), Leydig cells (LC) and endothelial cells of the blood vessels (BV). Female reproductive system and fertility The female reproductive system is a functional unit dedicated to the production of gametes, sperm transport and sperm–oocyte interaction leading to fertilisation, embryo–fetal development and delivery. It is composed of various elements with functional and endocrinologic interdependence on one another. It is characterised by cyclicity, i.e. under the initiation and regulation by the hypothalamic–pituitary–ovarian (HPO) axis, the reproductive tract shows well defined and sequential functional and morphological modifications. Elements of anatomy Anatomically, the female reproductive system includes the ovaries, oviducts or Fallopian tubes, uterus, cervix and vagina. The ovary is the primary reproductive gland, producing the female gametes (eggs or ova or oocytes); it is paired and situated within the pelvic cavity on either side adjacent to the uterus. The outer portion of the ovary, the cortex, contains the ovarian follicles and corpora lutea, the inner or central portion, the medulla, contains interstitial glands, larger calibre blood vessels and nerve fibres. Ovarian follicles can be morphologically differentiated in primordial, primary, secondary, tertiary follicles and preovulatory (Graafian) follicles, representing different stages of maturation (see below). Following ovulation, the released oocyte moves to the uterus through the oviducts or fallopian tubes (Avilés et al., 2015). These are paired tubes composed of the infundibulum (adjacent to the ovary, including the fimbriae), the ampulla and the isthmus (which joins the uterus). The fimbriae and infundibulum collect the oocyte and convey it to the ampulla where fertilisation typically occurs. Oviduct mucosa includes ciliated and non‐ciliated cells, along with mucous glands that facilitate the movement of the (fertilised) ovum towards the uterus and its nourishment. The uterus is a fibromuscular organ, with variable shape and dimensions (weight and size) across the species and subjected to change during pregnancy and hormonal fluctuations. It is composed of three layers: the endometrium, lining the luminal surface and composed of a mucosa organising in tubular glands embedded in a specialised and highly cellular stroma; the myometrium, a smooth muscle layer with inner circular and outer longitudinal fibres; and the adventitia (Jones & Lopez, 2014). Via the cervix, the uterus connects to the vagina, a distensible, thin‐walled tubular structure with a mucosa composed of a stratified epithelium, which undergoes changes across the oestrous cycle, serving as a reliable histologic indicator of the stage of the cycle in rodents (Sato et al., 2016; Westwood, 2008). The vulva is the external portion of the genitalia and surrounds the vaginal opening. Follicular development in the ovary begins during the fetal stage when germ cells transform into oogonia, which develop into primordial follicles. These follicles remain dormant until puberty, when the pituitary gland secretes FSH and luteinising hormone (LH), triggering them to mature into primary and then secondary follicles. During this process, granulosa cells proliferate, and fluid‐filled cavities called antra begin to form, increasing oestrogen production. The follicles continue to grow into antral follicles, and eventually, the preovulatory or Graafian follicle reaches its peak size, leading to an LH surge that triggers ovulation. Post‐ovulation, the follicle transforms into the corpus luteum, which secretes hormones to prepare the endometrium for potential pregnancy. If fertilisation does not occur, the corpus luteum degenerates into the corpus albicans, marking the end of the cycle. Fertilisation and pregnancy If sperm is present in the fallopian tube and successfully fertilises the egg, the resulting zygote undergoes multiple divisions while progressing through the fallopian tube towards the uterus, whereas the blastocyst is prepared for embedding into the uterine mucosa (implantation). Successful implantation leads to pregnancy, where the blastocyst matures into an embryo and eventually a fetus. In the absence of fertilisation, the egg proceeds through the fallopian tube to reach the uterus, where in humans and non‐human primates, it disintegrates and sheds along with the uterine lining during menstruation. Cyclicity and Hormonal Regulation of the Female Reproductive System (Boyd et al., 2018; Haschek et al., 2010; OECD GD No. 106, 2009; Vogazianou, 2019) The female reproductive system shows cyclic functional and morphological changes (cycles); each cycle begins with follicular growth and maturation, followed by ovulation (release of the ovum from the follicle), and the subsequent formation and regression of the corpus luteum. Cycling is regulated by the HPO axis, which coordinates hormonal signals through mechanisms of positive or negative feedback. The cyclic changes within the ovary are mainly regulated by LH and FSH, whereas the cyclic changes of the uterus and vagina are dependent on ovarian steroids. The pulsatile secretion of GnRH from the hypothalamus stimulates the anterior pituitary to secrete FSH and LH, which in turn stimulate the development of ovarian follicles and the production of ovarian steroids. FSH stimulates the development of the follicular zona granulosa and triggers the expression of LH receptors by granulosa cells. LH stimulates the synthesis and secretion of androstenedione and, to a lesser extent, testosterone by the theca interna of the follicle; these androgens are utilised by granulosa cells as substrates in the synthesis of oestrogen. The rise in oestrogen levels suppresses the release of GnRH by the hypothalamus, as well as directly inhibits pituitary secretion of both LH and FSH. Negative feedback control of pituitary FSH secretion is also achieved by the peptide inhibin, produced by the granulosa cells of the maturing follicle as depicted in Figure 4. Declining FSH levels cause follicular atresia of all the recruited follicles, except for the dominant follicle. Once peak oestrogen levels are reached, GnRH and gonadotrophin secretion cease. At this point, oestrogen starts promoting again both hypothalamic GnRH release and anterior pituitary responsiveness to GnRH. This positive oestrogenic modulation of hypothalamic–pituitary function results in a preovulatory luteinising hormone‐releasing hormone (LHRH) surge and corresponding surge in LH which results in ovulation. Following ovulation, the follicular granulosa and thecal cells undergo luteinisation to form the corpus luteum. This structure functions for a specific period known as pseudopregnancy. The corpus luteum initially secretes progesterone independently but requires stimulation by luteotropic hormones like LH and prolactin for sustained activity. Prolactin disrupts normal oestrus cycles by supporting the functional state of the corpus luteum. Progesterone secretion by the corpus luteum is crucial for endometrial development in preparation for potential embryo implantation and maintains vaginal mucification. If fertilisation does not occur, the corpus luteum regresses, allowing the return to normal reproductive cyclicity. The oestrous cycle in rodents and the menstrual cycle in humans represent fundamental differences in reproductive physiology. Rodents, such as rats, experience a relatively short and frequent oestrous cycle lasting around 4–6 days. This cycle consists of distinct phases including pro‐oestrus, oestrus, metoestrus and dioestrus (see Table 1). During this cycle, multiple follicles mature simultaneously and ovulation results in the release of multiple ova, allowing for potential multiple pregnancies. In contrast, the menstrual cycle in humans spans approximately 28 days, characterised by phases of menstrual shedding, follicular (proliferative) and luteal (secretory) phases. While multiple follicles start to mature in the human ovary, typically only one dominant follicle fully matures and releases an ovum per cycle. This singular ovulation pattern aligns with the human reproductive strategy, emphasising a higher investment in each pregnancy and a longer reproductive interval compared to rodents (Boyd et al., 2018). (A, B) The hypothalamic–pituitary–ovarian (HPO) axis and its role in female reproductive regulation. TABLE 1Key histological features of the reproductive organs at various stages of the oestrous cycle in rodents. For more details, see (Dixon et al., 2014; OECD GD No. 106, 2009; Sato et al., 2016; Westwood, 2008).Pro‐oestrusOestrusMetoestrusDioestrusOvary Large tertiary follicles are prominent during pro‐oestrus and located near the ovary's surface. Many antral tertiary follicles may be atretic. Following the peak in oestrogen, a luteinising hormone (LH) surge occurs, leading to a decline in oestrogen levels as late tertiary follicles begin to produce progesterone in preparation for ovulation. The corpus luteum from the last ovulation is a large degenerate with vacuolated luteal cells and fibrous tissue proliferation in the central cavity. There is an absence of healthy (non‐atretic) tertiary follicles. Several smaller vesicular follicles start to grow. Immature basophilic corpora lutea formed after the current ovulation have a central fluid‐filled cavity and no significant fibrous tissue proliferation. There are no healthy large late tertiary follicles, but many growing follicles of various types are present. The corpus luteum is characterised by a central fluid‐filled cavity, slightly basophilic cells and no fibrous tissue. Tertiary follicles increase in number but are smaller than those observed at pro‐oestrus. A large corpus luteum with fibrous tissue formation in the central cavity is observed. UterusThe epithelium ranges from cuboidal to columnar with noticeable mitotic figures. There is slight infiltration of inflammatory cells, and dilatation is observed towards the end of pro‐oestrus.This stage shows significant degeneration and necrosis of epithelial cells, loss of mitotic activity and leucocyte infiltration. Dilatation may persist into late oestrus.The endometrial epithelial cells continue to degenerate simultaneously with mitotic activity and degeneration.A small, avascular, slit‐like lumen is present. The epithelium is cuboidal to low columnar with few mitotic figures, which increase as the cycle progresses.VaginaMitotic figures, occasional polymorphs and minimal degeneration or desquamation are observed. The stratum granulosum develops, forming a superficial mucoid layer and corneum. Towards the end of this phase, keratinised stratum corneum is formed, and cornified cells become prominent, leading to some desquamation of mucoid cells.There is a gradual shedding of superficial mucoid and cornified layers, a reduction in epithelial height, the appearance of cell debris, loss of mitotic figures and gradual leucocyte infiltration.The cornified layer completely detaches initially, with persistent leucocyte infiltration, continued desquamation and loss of the stratum granulosum and upper germinativum.The onset is marked by variable leucocyte infiltration in the epithelium, followed by epithelial proliferation and thickening. Leucocyte infiltration ceases later in this phase. Key histological features of the reproductive organs at various stages of the oestrous cycle in rodents. For more details, see (Dixon et al., 2014; OECD GD No. 106, 2009; Sato et al., 2016; Westwood, 2008). Large tertiary follicles are prominent during pro‐oestrus and located near the ovary's surface. Many antral tertiary follicles may be atretic. Following the peak in oestrogen, a luteinising hormone (LH) surge occurs, leading to a decline in oestrogen levels as late tertiary follicles begin to produce progesterone in preparation for ovulation. The corpus luteum from the last ovulation is a large degenerate with vacuolated luteal cells and fibrous tissue proliferation in the central cavity. There is an absence of healthy (non‐atretic) tertiary follicles. Several smaller vesicular follicles start to grow. Immature basophilic corpora lutea formed after the current ovulation have a central fluid‐filled cavity and no significant fibrous tissue proliferation. There are no healthy large late tertiary follicles, but many growing follicles of various types are present. The corpus luteum is characterised by a central fluid‐filled cavity, slightly basophilic cells and no fibrous tissue. Tertiary follicles increase in number but are smaller than those observed at pro‐oestrus. A large corpus luteum with fibrous tissue formation in the central cavity is observed. Chemically induced alteration of the reproductive function including fertility Exposure to toxicants can have a substantial impact on reproductive function by disrupting various sites, including the hypothalamus, pituitary gland and male and female reproductive organs/systems. These disruptions can occur through direct and indirect mechanisms and an effect on one site often leads to subsequent changes at other sites. Therefore, it is essential to comprehensively examine a broad range of endpoints, such as organ weights, histopathology, gamete parameters, hormonal measurements and fertility endpoints. This comprehensive approach increases the likelihood of identifying toxicity, distinguishing primary toxic effects from secondary consequences, and enhances the ability to detect isolated random changes that may not have biological significance. Males Chemical exposure may impact testicular, epididymal and male accessory sex gland morphology and/or function. Impact of chemical exposure on testes: Testicular toxicants can target specific cellular components within the testes (germ cells, Sertoli cells, Leydig cells as well as interstitial macrophages and blood vessels), with a common outcome being the disruption of the process of spermatogenesis. Germ cells can undergo degeneration and depletion through direct cytotoxic effects or hormonal effects. The distribution pattern of the stage and germ cell type affected may be used to infer the mechanism of the injury, and a careful examination of the seminiferous epithelium can provide valuable information. Cytotoxicity typically targets proliferating spermatogonia, mitotically active and unprotected by the blood‐testis barrier (e.g. ethane‐methane sulfonate causes necrosis of spermatogonia by irreversible DNA damage). Hormonal disruption can lead to a decrease in the production of androgens (such as testosterone) preventing the androgen‐dependent development of germ cells. Such condition can occur following the administration of oestrogens or high doses of testosterone (Greaves, 2012; OECD GD No. 106, 2009). Effects at the level of the hypothalamus and pituitary gland (e.g. determining LH suppression affecting Leydig cell function) or androgen receptor blockade can also induce decreased testosterone levels/effects. In the spermatogenic cycle, Stages IX–XII and Stages VII/VIII are more sensitive to the level of testosterone. Early effects of androgen deficiency are recognised by degeneration of round spermatids and pachytene spermatocytes in stage VII/VIII tubules of rats. Long‐standing conditions can result in overall depletion of the seminiferous epithelium and tubular atrophy. Sertoli cell toxicity manifests primarily through loss of function, as these cells are resistant to death, or morphologically as vacuolation. The molecular mechanisms underlying these changes include disruptions in cytoskeletal function, metabolic injury resulting in decreased metabolic support of germ cells and effects on Leydig cell steroidogenesis or paracrine factors, which secondarily affect Sertoli cell function (Creasy et al., 2012; Woldemeskel, 2017). Subsequently, germ cells undergo multifocal degeneration with the formation of multinucleate spermatid aggregates, retention of spermatids /delayed spermiation (e.g. Boric acid, dichloroacetic acid and bromoacetic acid), phagocytosis of spermatocytes (e.g. 1,3‐Dinitrobenzene tri‐o‐cresylphosphate) and exfoliation of germ cells into the seminiferous tubular lumen associated with spermatogenic disruption in the testis (e.g. sloughing of the adluminal portion of Sertoli cells along with their attached germ cells induced by colchicine or vinblastine). With prolonged exposure at sufficiently high doses of toxicant, the tubules may undergo atrophy with seminiferous epithelium's disorganisation and germ cell loss, lined only by Sertoli cells. If the injury is sufficiently severe or prolonged, Sertoli cell function is compromised and spermatogenesis is not possible any longer (Creasy et al., 2012). Other effects leading to impaired spermatogenesis can include fluid disturbances. Impaired fluid reabsorption in the efferent ducts/caput epididymis, distal obstruction in the excurrent duct system (rete testis, efferent ducts or initial segment), increased secretion of seminiferous tubule fluid by Sertoli cells or inhibition of peritubular contractions can result in increased seminiferous tubular fluid and fluid backpressure to the seminiferous epithelium, consequently leading to germ cell degeneration. Inadequate fluid reabsorption is observed e.g. following the administration of endothelin antagonists, the fungicide carbendazim and a 5HT agonist. On the contrary, excessive fluid reabsorption can result in sperm stasis/sperm granuloma (e.g. a‐Chlorohydrin) and blockage (Creasy et al., 2012; Creasy & Chapin, 2018; Piner et al., 2002). Vascular damage within the testis can lead to ischaemic injury varying in severity, depending on the compound involved and exposure conditions. Vasoactive compounds such as serotonin, histamine and epinephrine can induce vascular changes in the testis, typically resulting in less severe effects on germ cells and causing patchy tubular atrophy. In contrast, severe disturbances in the vascular endothelium, as caused by cadmium, impact all testicular components, including germ cells, Sertoli cells, Leydig cells, peritubular myoid cells and the interstitial vasculature. This vast damage leads to coagulative necrosis of all testicular structures and may be followed by secondary effects such as inflammation, fibrosis and mineralisation (Piner et al., 2002). Impact of chemical exposure on epididymis: The epididymis is a crucial organ for sperm maturation, and toxicants can significantly perturb its structure and function, leading to altered sperm maturation processes. To comprehensively demonstrate the impact of toxic insults on the epididymis, it is important to evaluate multiple epididymal segments (as lesions can be region‐specific, such as in the caput, corpus or cauda) and compartments (e.g. sperm vs. epithelium). Characterising toxicity in the epididymis is complex due to the potential for both direct and indirect alterations.Direct toxicants, such as methyl chloride and cyclophosphamide, can affect the structure and function of the epididymal interstitium and epithelium, altering luminal fluid composition and sperm quality. These chemicals can cause exfoliation of principal cells, changes in epithelial cell height in the caput and corpus regions, loss of clear cells in the proximal cauda, degeneration and necrosis (De Grava Kempinas and Klinefelter, 2015; Vidal & Whitney, 2014). Epididymal injury is commonly accompanied by inflammation, followed by the development of sperm granuloma (Creasy, 2001).Indirect toxic effects, such as androgen deprivation or secondary effects due to testicular toxicity, can impact the epididymis significantly. As an androgen‐dependent tissue, in the absence of adequate androgen level, the epididymis can experience several morphological changes such as decreased weight, presence of sloughed testicular germ cells in the epididymal lumen, reduced sperm content and ductal atrophy. Less commonly, epithelial single cell necrosis/apoptosis may occur in specific segments of the caput epididymis. This can result from the suppression of androgen synthesis in the testis, as seen with ethane methanesulfonate (EDS), or from direct inhibition of the androgen receptor in the epididymis, as with hydroxyflutamide (De Grava Kempinas & Klinefelter, 2015; Vidal & Whitney, 2014).However, both direct and indirect mechanisms can alter sperm transit and/or induce sperm stasis, impacting maturation or directly damaging sperm, compromising the epididymal sperm quantity and quality, such as sperm motility, sperm morphology and sperm composition. These endpoints contribute significantly to overall fertility (i.e. fertilising ability and normal embryo development; Klinefelter & Hess, 1998). Impact of chemical exposure on male accessory sex glands: The secretory function of seminal vesicles and prostate is androgen‐dependent and is susceptible to circulating testosterone concentrations. Decrease in organ weight with or without manifesting morphologically (from epithelium single cell necrosis/apoptosis to atrophy with reduction in acinar/vesicle lumina) and secretory activity can result from endocrine disruption mechanisms like a marked decrease in testosterone biosynthesis, LHRH receptor antagonism, androgen receptor antagonism and aromatase inhibition. Conversely, androgen receptor agonism may increase organ weight, possibly evident in increased acinar/vesicle size and secretion amount and epithelial hypertrophy/hyperplasia. Additionally, the secretory function may be affected by inhibitors of 5‐alpha reductase, which metabolises testosterone to dihydrotestosterone. As with Leydig cell atrophy, subtle changes can be difficult to recognise by qualitative morphologic evaluation, and organ weight is usually much more sensitive than histopathology for detecting decreased functionality activity at the endpoint. Worth to mention is inflammation, generally occurring as a response to tissue injury or necrosis associated with urogenital bacterial infections, that could impact on fertility (OECD GD No. 106, 2009). Impact of chemical exposure on testes: Testicular toxicants can target specific cellular components within the testes (germ cells, Sertoli cells, Leydig cells as well as interstitial macrophages and blood vessels), with a common outcome being the disruption of the process of spermatogenesis. Germ cells can undergo degeneration and depletion through direct cytotoxic effects or hormonal effects. The distribution pattern of the stage and germ cell type affected may be used to infer the mechanism of the injury, and a careful examination of the seminiferous epithelium can provide valuable information. Cytotoxicity typically targets proliferating spermatogonia, mitotically active and unprotected by the blood‐testis barrier (e.g. ethane‐methane sulfonate causes necrosis of spermatogonia by irreversible DNA damage). Hormonal disruption can lead to a decrease in the production of androgens (such as testosterone) preventing the androgen‐dependent development of germ cells. Such condition can occur following the administration of oestrogens or high doses of testosterone (Greaves, 2012; OECD GD No. 106, 2009). Effects at the level of the hypothalamus and pituitary gland (e.g. determining LH suppression affecting Leydig cell function) or androgen receptor blockade can also induce decreased testosterone levels/effects. In the spermatogenic cycle, Stages IX–XII and Stages VII/VIII are more sensitive to the level of testosterone. Early effects of androgen deficiency are recognised by degeneration of round spermatids and pachytene spermatocytes in stage VII/VIII tubules of rats. Long‐standing conditions can result in overall depletion of the seminiferous epithelium and tubular atrophy. Sertoli cell toxicity manifests primarily through loss of function, as these cells are resistant to death, or morphologically as vacuolation. The molecular mechanisms underlying these changes include disruptions in cytoskeletal function, metabolic injury resulting in decreased metabolic support of germ cells and effects on Leydig cell steroidogenesis or paracrine factors, which secondarily affect Sertoli cell function (Creasy et al., 2012; Woldemeskel, 2017). Subsequently, germ cells undergo multifocal degeneration with the formation of multinucleate spermatid aggregates, retention of spermatids /delayed spermiation (e.g. Boric acid, dichloroacetic acid and bromoacetic acid), phagocytosis of spermatocytes (e.g. 1,3‐Dinitrobenzene tri‐o‐cresylphosphate) and exfoliation of germ cells into the seminiferous tubular lumen associated with spermatogenic disruption in the testis (e.g. sloughing of the adluminal portion of Sertoli cells along with their attached germ cells induced by colchicine or vinblastine). With prolonged exposure at sufficiently high doses of toxicant, the tubules may undergo atrophy with seminiferous epithelium's disorganisation and germ cell loss, lined only by Sertoli cells. If the injury is sufficiently severe or prolonged, Sertoli cell function is compromised and spermatogenesis is not possible any longer (Creasy et al., 2012). Other effects leading to impaired spermatogenesis can include fluid disturbances. Impaired fluid reabsorption in the efferent ducts/caput epididymis, distal obstruction in the excurrent duct system (rete testis, efferent ducts or initial segment), increased secretion of seminiferous tubule fluid by Sertoli cells or inhibition of peritubular contractions can result in increased seminiferous tubular fluid and fluid backpressure to the seminiferous epithelium, consequently leading to germ cell degeneration. Inadequate fluid reabsorption is observed e.g. following the administration of endothelin antagonists, the fungicide carbendazim and a 5HT agonist. On the contrary, excessive fluid reabsorption can result in sperm stasis/sperm granuloma (e.g. a‐Chlorohydrin) and blockage (Creasy et al., 2012; Creasy & Chapin, 2018; Piner et al., 2002). Vascular damage within the testis can lead to ischaemic injury varying in severity, depending on the compound involved and exposure conditions. Vasoactive compounds such as serotonin, histamine and epinephrine can induce vascular changes in the testis, typically resulting in less severe effects on germ cells and causing patchy tubular atrophy. In contrast, severe disturbances in the vascular endothelium, as caused by cadmium, impact all testicular components, including germ cells, Sertoli cells, Leydig cells, peritubular myoid cells and the interstitial vasculature. This vast damage leads to coagulative necrosis of all testicular structures and may be followed by secondary effects such as inflammation, fibrosis and mineralisation (Piner et al., 2002). Impact of chemical exposure on testes: Testicular toxicants can target specific cellular components within the testes (germ cells, Sertoli cells, Leydig cells as well as interstitial macrophages and blood vessels), with a common outcome being the disruption of the process of spermatogenesis. Germ cells can undergo degeneration and depletion through direct cytotoxic effects or hormonal effects. The distribution pattern of the stage and germ cell type affected may be used to infer the mechanism of the injury, and a careful examination of the seminiferous epithelium can provide valuable information. Cytotoxicity typically targets proliferating spermatogonia, mitotically active and unprotected by the blood‐testis barrier (e.g. ethane‐methane sulfonate causes necrosis of spermatogonia by irreversible DNA damage). Hormonal disruption can lead to a decrease in the production of androgens (such as testosterone) preventing the androgen‐dependent development of germ cells. Such condition can occur following the administration of oestrogens or high doses of testosterone (Greaves, 2012; OECD GD No. 106, 2009). Effects at the level of the hypothalamus and pituitary gland (e.g. determining LH suppression affecting Leydig cell function) or androgen receptor blockade can also induce decreased testosterone levels/effects. In the spermatogenic cycle, Stages IX–XII and Stages VII/VIII are more sensitive to the level of testosterone. Early effects of androgen deficiency are recognised by degeneration of round spermatids and pachytene spermatocytes in stage VII/VIII tubules of rats. Long‐standing conditions can result in overall depletion of the seminiferous epithelium and tubular atrophy. Sertoli cell toxicity manifests primarily through loss of function, as these cells are resistant to death, or morphologically as vacuolation. The molecular mechanisms underlying these changes include disruptions in cytoskeletal function, metabolic injury resulting in decreased metabolic support of germ cells and effects on Leydig cell steroidogenesis or paracrine factors, which secondarily affect Sertoli cell function (Creasy et al., 2012; Woldemeskel, 2017). Subsequently, germ cells undergo multifocal degeneration with the formation of multinucleate spermatid aggregates, retention of spermatids /delayed spermiation (e.g. Boric acid, dichloroacetic acid and bromoacetic acid), phagocytosis of spermatocytes (e.g. 1,3‐Dinitrobenzene tri‐o‐cresylphosphate) and exfoliation of germ cells into the seminiferous tubular lumen associated with spermatogenic disruption in the testis (e.g. sloughing of the adluminal portion of Sertoli cells along with their attached germ cells induced by colchicine or vinblastine). With prolonged exposure at sufficiently high doses of toxicant, the tubules may undergo atrophy with seminiferous epithelium's disorganisation and germ cell loss, lined only by Sertoli cells. If the injury is sufficiently severe or prolonged, Sertoli cell function is compromised and spermatogenesis is not possible any longer (Creasy et al., 2012). Other effects leading to impaired spermatogenesis can include fluid disturbances. Impaired fluid reabsorption in the efferent ducts/caput epididymis, distal obstruction in the excurrent duct system (rete testis, efferent ducts or initial segment), increased secretion of seminiferous tubule fluid by Sertoli cells or inhibition of peritubular contractions can result in increased seminiferous tubular fluid and fluid backpressure to the seminiferous epithelium, consequently leading to germ cell degeneration. Inadequate fluid reabsorption is observed e.g. following the administration of endothelin antagonists, the fungicide carbendazim and a 5HT agonist. On the contrary, excessive fluid reabsorption can result in sperm stasis/sperm granuloma (e.g. a‐Chlorohydrin) and blockage (Creasy et al., 2012; Creasy & Chapin, 2018; Piner et al., 2002). Vascular damage within the testis can lead to ischaemic injury varying in severity, depending on the compound involved and exposure conditions. Vasoactive compounds such as serotonin, histamine and epinephrine can induce vascular changes in the testis, typically resulting in less severe effects on germ cells and causing patchy tubular atrophy. In contrast, severe disturbances in the vascular endothelium, as caused by cadmium, impact all testicular components, including germ cells, Sertoli cells, Leydig cells, peritubular myoid cells and the interstitial vasculature. This vast damage leads to coagulative necrosis of all testicular structures and may be followed by secondary effects such as inflammation, fibrosis and mineralisation (Piner et al., 2002). Impact of chemical exposure on epididymis: The epididymis is a crucial organ for sperm maturation, and toxicants can significantly perturb its structure and function, leading to altered sperm maturation processes. To comprehensively demonstrate the impact of toxic insults on the epididymis, it is important to evaluate multiple epididymal segments (as lesions can be region‐specific, such as in the caput, corpus or cauda) and compartments (e.g. sperm vs. epithelium). Characterising toxicity in the epididymis is complex due to the potential for both direct and indirect alterations.Direct toxicants, such as methyl chloride and cyclophosphamide, can affect the structure and function of the epididymal interstitium and epithelium, altering luminal fluid composition and sperm quality. These chemicals can cause exfoliation of principal cells, changes in epithelial cell height in the caput and corpus regions, loss of clear cells in the proximal cauda, degeneration and necrosis (De Grava Kempinas and Klinefelter, 2015; Vidal & Whitney, 2014). Epididymal injury is commonly accompanied by inflammation, followed by the development of sperm granuloma (Creasy, 2001).Indirect toxic effects, such as androgen deprivation or secondary effects due to testicular toxicity, can impact the epididymis significantly. As an androgen‐dependent tissue, in the absence of adequate androgen level, the epididymis can experience several morphological changes such as decreased weight, presence of sloughed testicular germ cells in the epididymal lumen, reduced sperm content and ductal atrophy. Less commonly, epithelial single cell necrosis/apoptosis may occur in specific segments of the caput epididymis. This can result from the suppression of androgen synthesis in the testis, as seen with ethane methanesulfonate (EDS), or from direct inhibition of the androgen receptor in the epididymis, as with hydroxyflutamide (De Grava Kempinas & Klinefelter, 2015; Vidal & Whitney, 2014).However, both direct and indirect mechanisms can alter sperm transit and/or induce sperm stasis, impacting maturation or directly damaging sperm, compromising the epididymal sperm quantity and quality, such as sperm motility, sperm morphology and sperm composition. These endpoints contribute significantly to overall fertility (i.e. fertilising ability and normal embryo development; Klinefelter & Hess, 1998). Impact of chemical exposure on epididymis: The epididymis is a crucial organ for sperm maturation, and toxicants can significantly perturb its structure and function, leading to altered sperm maturation processes. To comprehensively demonstrate the impact of toxic insults on the epididymis, it is important to evaluate multiple epididymal segments (as lesions can be region‐specific, such as in the caput, corpus or cauda) and compartments (e.g. sperm vs. epithelium). Characterising toxicity in the epididymis is complex due to the potential for both direct and indirect alterations. Direct toxicants, such as methyl chloride and cyclophosphamide, can affect the structure and function of the epididymal interstitium and epithelium, altering luminal fluid composition and sperm quality. These chemicals can cause exfoliation of principal cells, changes in epithelial cell height in the caput and corpus regions, loss of clear cells in the proximal cauda, degeneration and necrosis (De Grava Kempinas and Klinefelter, 2015; Vidal & Whitney, 2014). Epididymal injury is commonly accompanied by inflammation, followed by the development of sperm granuloma (Creasy, 2001). Indirect toxic effects, such as androgen deprivation or secondary effects due to testicular toxicity, can impact the epididymis significantly. As an androgen‐dependent tissue, in the absence of adequate androgen level, the epididymis can experience several morphological changes such as decreased weight, presence of sloughed testicular germ cells in the epididymal lumen, reduced sperm content and ductal atrophy. Less commonly, epithelial single cell necrosis/apoptosis may occur in specific segments of the caput epididymis. This can result from the suppression of androgen synthesis in the testis, as seen with ethane methanesulfonate (EDS), or from direct inhibition of the androgen receptor in the epididymis, as with hydroxyflutamide (De Grava Kempinas & Klinefelter, 2015; Vidal & Whitney, 2014). However, both direct and indirect mechanisms can alter sperm transit and/or induce sperm stasis, impacting maturation or directly damaging sperm, compromising the epididymal sperm quantity and quality, such as sperm motility, sperm morphology and sperm composition. These endpoints contribute significantly to overall fertility (i.e. fertilising ability and normal embryo development; Klinefelter & Hess, 1998). Impact of chemical exposure on male accessory sex glands: The secretory function of seminal vesicles and prostate is androgen‐dependent and is susceptible to circulating testosterone concentrations. Decrease in organ weight with or without manifesting morphologically (from epithelium single cell necrosis/apoptosis to atrophy with reduction in acinar/vesicle lumina) and secretory activity can result from endocrine disruption mechanisms like a marked decrease in testosterone biosynthesis, LHRH receptor antagonism, androgen receptor antagonism and aromatase inhibition. Conversely, androgen receptor agonism may increase organ weight, possibly evident in increased acinar/vesicle size and secretion amount and epithelial hypertrophy/hyperplasia. Additionally, the secretory function may be affected by inhibitors of 5‐alpha reductase, which metabolises testosterone to dihydrotestosterone. As with Leydig cell atrophy, subtle changes can be difficult to recognise by qualitative morphologic evaluation, and organ weight is usually much more sensitive than histopathology for detecting decreased functionality activity at the endpoint. Worth to mention is inflammation, generally occurring as a response to tissue injury or necrosis associated with urogenital bacterial infections, that could impact on fertility (OECD GD No. 106, 2009). Impact of chemical exposure on male accessory sex glands: The secretory function of seminal vesicles and prostate is androgen‐dependent and is susceptible to circulating testosterone concentrations. Decrease in organ weight with or without manifesting morphologically (from epithelium single cell necrosis/apoptosis to atrophy with reduction in acinar/vesicle lumina) and secretory activity can result from endocrine disruption mechanisms like a marked decrease in testosterone biosynthesis, LHRH receptor antagonism, androgen receptor antagonism and aromatase inhibition. Conversely, androgen receptor agonism may increase organ weight, possibly evident in increased acinar/vesicle size and secretion amount and epithelial hypertrophy/hyperplasia. Additionally, the secretory function may be affected by inhibitors of 5‐alpha reductase, which metabolises testosterone to dihydrotestosterone. As with Leydig cell atrophy, subtle changes can be difficult to recognise by qualitative morphologic evaluation, and organ weight is usually much more sensitive than histopathology for detecting decreased functionality activity at the endpoint. Worth to mention is inflammation, generally occurring as a response to tissue injury or necrosis associated with urogenital bacterial infections, that could impact on fertility (OECD GD No. 106, 2009). Females Mechanistically, chemicals can induce direct effects and indirect effects on the female reproductive system. Direct effects include cytotoxicity to oocytes and/or granulosa cells [e.g. heavy metals, industrial chemicals (see Junaid et al., 1997)] and irradiation affecting small follicles, cisplatin and cyclophosphamide affecting medium or large antral follicles (see Plowchalk & Mattison, 1992). Also, direct effects are oestrogen‐ or progesterone‐like activity, such as exaggerated effects of endogenous 17b‐estradiol (E2) or progesterone (P4) due to increased serum concentrations or an imbalance of these two hormones. Morphological changes in this case can include atrophic or hypertrophic/hyperplastic changes, compatible with the induced changes in the hormonal milieu (see below) (Hooser et al., 1994). Inhibition of angiogenesis can also determine morphologic and functional alterations in the corpus luteum (decreased or increased in number, with central haemorrhage and degeneration) with secondary effects in the uterus and vagina (Hall et al., 2016, Patyna et al., 2008). Indirect effects are in most cases due to the perturbation of the hormonal pathway and can be mediated via interference of hypothalamic–pituitary pathways, manifesting as ovarian changes with downstream effects on the reproductive tract (and mammary gland). Chemically induced hormonal perturbations can manifest morphologically according to three main patterns (OECD GD No. 106, 2009): Atrophy of the ovary, uterus and vagina: It can be observed in the case of compounds exerting an antiestrogenic effect, or it can be caused by decreased gonadotropin production in the pituitary gland, decreased follicular development in the ovary or decreased ovarian steroidogenesis. Aromatase inhibitors are known compounds inducing such conditions. They can impair the conversion of androgens to oestrogens by the zona granulosa/theca cells of the ovarian follicles, with consequent reduced endogenous production of ovarian oestrogens and widespread histopathological atrophic changes (ovaries, uterus and vagina). In case of reduced production of ovarian oestrogens, impaired positive feedback to HPO could cause a disruption of the preovulatory LH source with ovarian increased follicular atresia and follicular cysts. Interstitial cell hypertrophy and hyperplasia may also occur following the administration of anti‐oestrogens due to increased levels of gonadotropins (Yuan & Foley, 2002). Ovarian atrophy, uterine and vaginal hypertrophy and hyperplasia: It can be noted in the case of compounds exerting oestrogenic effect. Such compounds could on one side determine negative feedback at the hypothalamic–pituitary level, with inhibition of the release of gonadotrophins and consequent ovarian atrophy; on the other side, they can stimulate the uterus and vagina, with organ hypertrophy/hyperplasia. Histopathological findings include ovarian atrophy, hypertrophy of the uterine luminal and/or glandular epithelium, myometrial hypertrophy and increased vaginal epithelial thickness with hyperkeratinisation (Biegel et al., 1998; Yamasaki et al., 2002). Compounds with oestrogenic effects include ethinyl oestradiol (oestrogen agonist) and tamoxifen/raloxifene (selective oestrogen receptor modulators) (OECD GD No. 106, 2009). Oestradiol valerate, a synthetic steroidal oestrogen, can also induce follicular cysts in association with high FSH levels and low LH (Grosser et al., 1987). Hypertrophy/hyperplasia of the vagina, uterus and ovary due to increased pituitary hormone concentrations. This can be noted with compounds with a gonadotropic or luteotropic effect. High gonadotropic levels trigger continuous follicle maturation and corpora lutea formation, increased ovarian activity with elevated oestrogen and progesterone levels and proliferative changes in the uterus and vagina. In the vagina, increased and persistent mucification can occur. High prolactin (PRL) levels lead to increased size and/or number of corpora lutea (persistent corpora lutea e.g. in mice and rats) and mammary gland lobuloalveolar hyperplasia. Example substances are analogues of LH, LHRH, FSH and prolactin, and dopamine‐depleting compounds (e.g. reserpine). Ethylene glycol monomethyl ether (EGME) also promotes luteal survival and progesterone production indirectly through upregulation of prolactin secretion (Taketa et al., 2011). Atrophy of the ovary, uterus and vagina: It can be observed in the case of compounds exerting an antiestrogenic effect, or it can be caused by decreased gonadotropin production in the pituitary gland, decreased follicular development in the ovary or decreased ovarian steroidogenesis. Aromatase inhibitors are known compounds inducing such conditions. They can impair the conversion of androgens to oestrogens by the zona granulosa/theca cells of the ovarian follicles, with consequent reduced endogenous production of ovarian oestrogens and widespread histopathological atrophic changes (ovaries, uterus and vagina). In case of reduced production of ovarian oestrogens, impaired positive feedback to HPO could cause a disruption of the preovulatory LH source with ovarian increased follicular atresia and follicular cysts. Interstitial cell hypertrophy and hyperplasia may also occur following the administration of anti‐oestrogens due to increased levels of gonadotropins (Yuan & Foley, 2002). Atrophy of the ovary, uterus and vagina: It can be observed in the case of compounds exerting an antiestrogenic effect, or it can be caused by decreased gonadotropin production in the pituitary gland, decreased follicular development in the ovary or decreased ovarian steroidogenesis. Aromatase inhibitors are known compounds inducing such conditions. They can impair the conversion of androgens to oestrogens by the zona granulosa/theca cells of the ovarian follicles, with consequent reduced endogenous production of ovarian oestrogens and widespread histopathological atrophic changes (ovaries, uterus and vagina). In case of reduced production of ovarian oestrogens, impaired positive feedback to HPO could cause a disruption of the preovulatory LH source with ovarian increased follicular atresia and follicular cysts. Interstitial cell hypertrophy and hyperplasia may also occur following the administration of anti‐oestrogens due to increased levels of gonadotropins (Yuan & Foley, 2002). Ovarian atrophy, uterine and vaginal hypertrophy and hyperplasia: It can be noted in the case of compounds exerting oestrogenic effect. Such compounds could on one side determine negative feedback at the hypothalamic–pituitary level, with inhibition of the release of gonadotrophins and consequent ovarian atrophy; on the other side, they can stimulate the uterus and vagina, with organ hypertrophy/hyperplasia. Histopathological findings include ovarian atrophy, hypertrophy of the uterine luminal and/or glandular epithelium, myometrial hypertrophy and increased vaginal epithelial thickness with hyperkeratinisation (Biegel et al., 1998; Yamasaki et al., 2002). Compounds with oestrogenic effects include ethinyl oestradiol (oestrogen agonist) and tamoxifen/raloxifene (selective oestrogen receptor modulators) (OECD GD No. 106, 2009). Oestradiol valerate, a synthetic steroidal oestrogen, can also induce follicular cysts in association with high FSH levels and low LH (Grosser et al., 1987). Ovarian atrophy, uterine and vaginal hypertrophy and hyperplasia: It can be noted in the case of compounds exerting oestrogenic effect. Such compounds could on one side determine negative feedback at the hypothalamic–pituitary level, with inhibition of the release of gonadotrophins and consequent ovarian atrophy; on the other side, they can stimulate the uterus and vagina, with organ hypertrophy/hyperplasia. Histopathological findings include ovarian atrophy, hypertrophy of the uterine luminal and/or glandular epithelium, myometrial hypertrophy and increased vaginal epithelial thickness with hyperkeratinisation (Biegel et al., 1998; Yamasaki et al., 2002). Compounds with oestrogenic effects include ethinyl oestradiol (oestrogen agonist) and tamoxifen/raloxifene (selective oestrogen receptor modulators) (OECD GD No. 106, 2009). Oestradiol valerate, a synthetic steroidal oestrogen, can also induce follicular cysts in association with high FSH levels and low LH (Grosser et al., 1987). Hypertrophy/hyperplasia of the vagina, uterus and ovary due to increased pituitary hormone concentrations. This can be noted with compounds with a gonadotropic or luteotropic effect. High gonadotropic levels trigger continuous follicle maturation and corpora lutea formation, increased ovarian activity with elevated oestrogen and progesterone levels and proliferative changes in the uterus and vagina. In the vagina, increased and persistent mucification can occur. High prolactin (PRL) levels lead to increased size and/or number of corpora lutea (persistent corpora lutea e.g. in mice and rats) and mammary gland lobuloalveolar hyperplasia. Example substances are analogues of LH, LHRH, FSH and prolactin, and dopamine‐depleting compounds (e.g. reserpine). Ethylene glycol monomethyl ether (EGME) also promotes luteal survival and progesterone production indirectly through upregulation of prolactin secretion (Taketa et al., 2011). Hypertrophy/hyperplasia of the vagina, uterus and ovary due to increased pituitary hormone concentrations. This can be noted with compounds with a gonadotropic or luteotropic effect. High gonadotropic levels trigger continuous follicle maturation and corpora lutea formation, increased ovarian activity with elevated oestrogen and progesterone levels and proliferative changes in the uterus and vagina. In the vagina, increased and persistent mucification can occur. High prolactin (PRL) levels lead to increased size and/or number of corpora lutea (persistent corpora lutea e.g. in mice and rats) and mammary gland lobuloalveolar hyperplasia. Example substances are analogues of LH, LHRH, FSH and prolactin, and dopamine‐depleting compounds (e.g. reserpine). Ethylene glycol monomethyl ether (EGME) also promotes luteal survival and progesterone production indirectly through upregulation of prolactin secretion (Taketa et al., 2011). Administration of compounds with hormonal activity, particularly oestrogens, may also induce uterine inflammation, such as endometritis and pyometra (Sahota et al., 2018). Further details are reported in Appendix C. These morphological changes can affect fertility/reproductive functional endpoints very broadly. In addition, care should be taken in interpreting fertility/reproductive functional endpoints in the presence of excessive toxicity. Regulatory assessment of pesticides for effects on reproduction and fertility Test guidelines of reproductive function and fertility in regulatory settings In the regulatory setting, reproductive function is largely assessed in reproductive toxicity studies provided to fulfil data requirements described in the legal text (e.g. Regulation (EC) 1107/2009, Regulation (EU) 283/2013, Regulation (EU) 528/2012), supported by all other available information including physicochemical properties, toxicokinetic data, in silico predictions and data on structural analogues, in vitro and in vivo animal data published in scientific literature, other general toxicity studies, medical records and epidemiological data. Reproductive toxicity studies may include generational studies (OECD Test Guideline (TG) 415, OECD TG 416, OECD TG 443) as well as reproduction screening tests (OECD TG 421, OECD TG 422). An overview of the objectives concerning integrity and performance of adult male and female reproductive systems of the most recent versions of OECD TGs on reproductive toxicity is included in Table 2. TABLE 2Outline of the main objectives of OECD Test Guidelines concerning reproductive function and fertility.OECD test guideline (last update) a Type of testObjective concerning integrity and performance of adult male and female reproductive systemsOECD TG 415 (1983)One‐generation reproduction toxicity studyGonadal function, oestrous cycle, mating behaviour, conception, parturitionOECD TG 416 (2001)Two‐generation reproduction toxicity studyGonadal function, oestrus cycle, mating behaviour, conception, gestation, parturitionOECD TG 421 (2016) b Reproduction/developmental toxicity screening testGonadal function, mating behaviour, conception and parturitionOECD TG 422 (2016) b Combined repeated dose toxicity study with the reproduction/developmental toxicity screening testOECD TG 443 (2018)Extended one‐generation reproductive toxicity studyGonadal function, oestrous cycle, epididymal sperm maturation, mating behaviour, conception, pregnancy and parturition. a Older pesticide dossiers may only include one two‐generation reproductive toxicity study conducted according to the OECD TG 416 from 1983 where several parameters are not investigated (differences in collection of parameters between OECD 416 versions are explained in detail in Table 3 (males) and 6 (females)). b These methods provide limited information on effects on male and female reproductive function and as such they will not provide evidence for definite claims of no effects. In the absence of other data on reproductive toxicity, positive results are useful for initial hazard assessment and contribute to decisions with respect to the necessity and timing of additional testing. Outline of the main objectives of OECD Test Guidelines concerning reproductive function and fertility. Older pesticide dossiers may only include one two‐generation reproductive toxicity study conducted according to the OECD TG 416 from 1983 where several parameters are not investigated (differences in collection of parameters between OECD 416 versions are explained in detail in Table 3 (males) and 6 (females)). These methods provide limited information on effects on male and female reproductive function and as such they will not provide evidence for definite claims of no effects. In the absence of other data on reproductive toxicity, positive results are useful for initial hazard assessment and contribute to decisions with respect to the necessity and timing of additional testing. Other OECD guidance documents provide support on the design and assessment of reproductive toxicity studies: OECD (2008) Guidance Document on Mammalian Reproductive Toxicity Testing and Assessment (No. 43).OECD (2009) Guidance Document for Histologic Evaluation of Endocrine and Reproductive Tests in Rodents (No. 106).OECD (2011) Guidance Document on Standardised Test Guidelines for Evaluating Chemicals for Endocrine Disruption (No. 150). OECD (2008) Guidance Document on Mammalian Reproductive Toxicity Testing and Assessment (No. 43). OECD (2008) Guidance Document on Mammalian Reproductive Toxicity Testing and Assessment (No. 43). OECD (2009) Guidance Document for Histologic Evaluation of Endocrine and Reproductive Tests in Rodents (No. 106). OECD (2009) Guidance Document for Histologic Evaluation of Endocrine and Reproductive Tests in Rodents (No. 106). OECD (2011) Guidance Document on Standardised Test Guidelines for Evaluating Chemicals for Endocrine Disruption (No. 150). OECD (2011) Guidance Document on Standardised Test Guidelines for Evaluating Chemicals for Endocrine Disruption (No. 150). In addition, more specific details on the implementation of the extended one‐generation reproductive toxicity (EOGRT) study under the OECD TG 443 are included in the OECD Guidance Document (GD) No. 117 concerning the Current Implementation of Internal Triggers in Test Guideline 443 for an Extended One Generation Reproductive Toxicity Study, in the United States and Canada, as well as in the OECD GD No. 151 on the design, conduction and interpretation of the OECD TG 443. The overall objective of testing for reproductive performance under European jurisdiction is to have adequate information for hazard identification, including the description of specific adverse effects on sexual function and fertility and classification and labelling proposals for reproductive effects (fertility), as well as hazard characterisation and the setting of reproductive toxicity reference points, i.e. reproductive NOAEL/LOAEL/BMDL. Furthermore, additional studies, e.g. OECD TG 408, are used in data collection since they are recognised as relevant to complement information from fertility and reproductive toxicity studies. Parameters of male reproductive function and fertility in regulatory toxicology An overview of the reproductive toxicity parameters investigated in current OECD TGs and considered in the regulatory assessment of male reproductive function and fertility is included in Table 3 and Table B.1 in Appendix B. TABLE 3Overview of reproductive toxicity parameters considered in regulatory assessment of male reproductive function and fertility.Reproductive toxicity parameters considered in regulatory assessment of reproductive function and fertility in male rodentsOECD test guideline (last update a )Endpoint indicative of ‘Sensitive to, but not diagnostic of, EAS’Mating performanceTime to mating (pre‐coital interval: time duration from pairing to insemination) 443 (2018) 416 (2001) 415 (1983) 421 (2016) 422 (2016) Number of successful matingsFertility (function)Number of males siring a litterEndpoint indicative of ‘EAS‐mediated’ effectFertility (Reproductive organ weight and histopathology b )Testis weight 443 (2018) 416 (2001) 421 (2016) 422 (2016) [407 (2008)] e 408 (2018) Epididymis weightProstate weightSeminal vesicles and coagulating gland weight (measured as one unit)Levator ani/bulbocavernosus muscle complex (LABC) weight 421 (2016) 422 (2016) Cowper's gland weightGlans penis weightTestis histopathology (detailed) 443 (2018) 416 (2001) 415 (1983) c 421 (2016) d 422 (2016) d [407 (2008)] e 408 (2018) Epididymis histopathologyProstate histopathologySeminal vesicles histopathologyCoagulating gland histopathologyFertility (Sperm parameters)Sperm morphology: ‐% sperm morphologically normal (both head and midpiece/tail appear normal),‐% sperm with each identified abnormality (type of abnormality: fusion, isolated heads and misshapen heads and/or tails) 443 (2018) 416 (2001) 408 (2018) Sperm motility (%)Sperm numbers: ‐testicular spermatids, and/or‐total cauda epididymal sperm reserves Abbreviations: EAS, Oestrogen, Androgen, Steroidogenesis endpoints as described in the ECHA/EFSA Guidance on the identification of endocrine disruptors in the context of Regulations (EU) No 528/2012 and (EC) No 1107/2009 (2018). a Older pesticide dossier may only include one two‐generation reproductive toxicity study conducted according to the OECD TG 416 from 1983 and several parameters are not investigated, i.e. reproductive organ weights and sperm parameters are not examined and histopathology of reproductive organs (testes, epididymis, seminal vesicles, prostate, coagulating gland) is only performed when necessary or if the organs have not been histologically examined in other multiple‐dose studies. b Histopathology is performed in male animals of the control and high‐dose groups. Examination in other dose groups is extended when changes are seen in the highest dose groups. c Male reproductive organ histopathology in OECD TG 415 (1983) is only performed when necessary (e.g. in case of macroscopic lesions). d In OECD TG 421 (2016) and 422 (2016), histopathology of the testes and epididymis should be considered along with the fertility data, when assessing male reproductive effects because of the short period of treatment of the male. Prostate, seminal vesicles and coagulating gland histopathology is only performed when necessary (e.g. in case of macroscopic lesions). Stages of spermatogenesis in the male gonads are assessed as part of tissue histopathology. e Dose‐range finding studies are used to provide information on dose/concentration selection for studies of longer duration. Overview of reproductive toxicity parameters considered in regulatory assessment of male reproductive function and fertility. 443 (2018) 416 (2001) 415 (1983) 421 (2016) 422 (2016) 443 (2018) 416 (2001) 421 (2016) 422 (2016) [407 (2008)] e 408 (2018) 421 (2016) 422 (2016) 443 (2018) 416 (2001) 415 (1983) c 421 (2016) d 422 (2016) d [407 (2008)] e 408 (2018) ‐% sperm morphologically normal (both head and midpiece/tail appear normal), % sperm morphologically normal (both head and midpiece/tail appear normal), ‐% sperm with each identified abnormality (type of abnormality: fusion, isolated heads and misshapen heads and/or tails) % sperm with each identified abnormality (type of abnormality: fusion, isolated heads and misshapen heads and/or tails) 443 (2018) 416 (2001) 408 (2018) ‐testicular spermatids, and/or testicular spermatids, and/or ‐total cauda epididymal sperm reserves total cauda epididymal sperm reserves Abbreviations: EAS, Oestrogen, Androgen, Steroidogenesis endpoints as described in the ECHA/EFSA Guidance on the identification of endocrine disruptors in the context of Regulations (EU) No 528/2012 and (EC) No 1107/2009 (2018). Older pesticide dossier may only include one two‐generation reproductive toxicity study conducted according to the OECD TG 416 from 1983 and several parameters are not investigated, i.e. reproductive organ weights and sperm parameters are not examined and histopathology of reproductive organs (testes, epididymis, seminal vesicles, prostate, coagulating gland) is only performed when necessary or if the organs have not been histologically examined in other multiple‐dose studies. Histopathology is performed in male animals of the control and high‐dose groups. Examination in other dose groups is extended when changes are seen in the highest dose groups. Male reproductive organ histopathology in OECD TG 415 (1983) is only performed when necessary (e.g. in case of macroscopic lesions). In OECD TG 421 (2016) and 422 (2016), histopathology of the testes and epididymis should be considered along with the fertility data, when assessing male reproductive effects because of the short period of treatment of the male. Prostate, seminal vesicles and coagulating gland histopathology is only performed when necessary (e.g. in case of macroscopic lesions). Stages of spermatogenesis in the male gonads are assessed as part of tissue histopathology. Dose‐range finding studies are used to provide information on dose/concentration selection for studies of longer duration. The number of males paired, the number of males with evidence of mating and the number of males siring a litter are used in the calculation of reproductive toxicity indices that are considered as measures of reproductive performance. As OECD test guidelines do not require it, the number of mounts is not usually collected, and therefore, it is often not known if prolonged mating is due to not pairing or unsuccessful pairing. If this information is available, this will be collected in the data collection for the indicator ‘delayed time to mating’ (see Section 4.1.1). Male reproductive indices are calculated for all dose levels in P generation animals (and also F1 generation animals depending on the study design) as indicated in the OECD Guidance Document No. 43 on mammalian reproductive toxicity testing and assessment (2008): TABLE 4Calculation of male reproductive indices.* Male reproductive indicesCalculationDefinitionMale mating index No.of males with confirmed matingTotalNo.of males cohabitedx100 Measure of male's ability to mateMale fertility index No.of males impregnatingafemaleTotalNo.of males cohabitedx100 Measure of male's ability to produce sperm that can fertilise eggs*Information extracted from Table 7 of the OECD GD No. 43 (2008). Calculation of male reproductive indices.* Information extracted from Table 7 of the OECD GD No. 43 (2008). Measurements of male reproductive hormone levels are not foreseen in the currently available OECD test guidelines on reproductive toxicity (Table 2) and repeated dose short‐term toxicity (OECD TG 407, 408, 409). Substances that affect testosterone production may be detected in the in vitro steroidogenesis assay performed according to the OECD TG 456 (2023). Additional information on a potential (anti)androgenic effect of a substance suggesting disruption of the male reproductive system through an endocrine mode of action may be generated in the in vitro OECD TG 458 (2023) androgen receptor transactivation assay, the ToxCast ΑR Bioactivity Model (US EPA CompTox Chemicals Dashboard) and the in vivo OECD TG 441 (2009) Hershberger bioassay in rats. These in vitro and in vivo assays (Table 5) that provide data on selected endocrine mechanism(s)/pathway(s) are described in the ECHA/EFSA Guidance on the identification of endocrine disruptors in the context of Regulations (EU) No 528/2012 and (EC) No 1107/2009 (2018). TABLE 5Recommended assays for the detection of androgen receptor (ant)agonists and substances affecting testosterone production.Assay familyOECD test guideline (last update)Type of assayAssay objectiveTransactivation assayOECD TG 458 (2023)Androgen receptor transactivation (ARTA) assayDetects agonists and antagonists of androgen receptor in vitro.OECD TG 441 (2009)Hershberger bioassay in rats: A short‐term screening assay for (anti)androgenic propertiesEvaluates the ability of a chemical to elicit biological activities consistent with androgen agonists, antagonists or 5α‐reductase inhibitors.SteroidogenesisOECD TG 456 (2023)* H295R steroidogenesis assayDetects substances that affect Testosterone production*Other validated Guideline: OPPTS 890.1550 (2011). Recommended assays for the detection of androgen receptor (ant)agonists and substances affecting testosterone production. Other validated Guideline: OPPTS 890.1550 (2011). Parameters of female reproductive function and fertility in regulatory toxicology An overview of the reproductive toxicity parameters investigated in current OECD TGs and considered in the regulatory assessment of female reproductive performance is included in Table 6 and Table B.2 in Appendix B. TABLE 6Overview of reproductive toxicity parameters considered in regulatory assessment of female reproductive function and fertility.Reproductive toxicity parameters considered in regulatory assessment of reproductive function and fertility in female rodentsOECD test guideline (last update a )Endpoint indicative of ‘Sensitive to, but not diagnostic of, EAS’Mating performanceTime to mating (pre‐coital interval: time duration from pairing to insemination) 443 (2018) 416 (2001) Number of successful matingsEndpoint indicative of ‘EAS‐mediated’ effectFertility (function)Number of normal or abnormal oestrus cycles by vaginal smears (during lifetime) 443 (2018) 416 (2001) 415 (1983) 421 (2016) 422 (2016) Oestrus cycle length (during lifetime)Endpoint indicative of ‘Sensitive to, but not diagnostic of, EAS’Number of pregnant femalesNumber of ovarian follicles 443 (2018) 416 (2001) 421 (2016) b 422 (2016) b Maturation stage of ovarian folliclesNumber of implantation sites f Number of corpora lutea in the ovariesEndpoint indicative of ‘EAS‐mediated’ effectFertility (reproductive organ weight, cytology and histopathology c )Ovary weight 443 (2018) 416 (2001) 421 (2016) d 422 (2016) d [407 (2008) d ] g 408 (2018) Uterus weightCervix weight 443 (2018) 421 (2016) d 422 (2016) d 407 (2008) d Ovary histopathology 443 (2018) 416 (2001) 415 (1983) e 421 (2016) e 422 (2016) [407 (2008)] g 408 (2018) Uterus histopathology (with cervix)Vagina histopathologyVaginal smears 443 (2018) 421 (2016) e 422 (2016) [407 (2008)] g 408 (2018) Oviduct histopathology443 (2018)Endpoint indicative of ‘Sensitive to, but not diagnostic of, EAS’GestationNumber of females delivering a viable litter 443 (2018) 416 (2001) 415 (1983) 421 (2016) 422 (2016) Duration of gestation (insemination to parturition)Endpoint indicative of ‘Sensitive to, but not diagnostic of, EAS’ParturitionNumber of females with Dystocia (prolonged or difficult parturition)Abbreviations: EAS, Oestrogen, Androgen, Steroidogenesis endpoints described in the ECHA/EFSA Guidance on the identification of endocrine disruptors in the context of Regulations (EU) No 528/2012 and (EC) No 1107/2009 (2018). a Older pesticide dossiers may only include one two‐generation reproductive toxicity study conducted according to the OECD TG 416 from 1983 and several parameters are not investigated, i.e. oestrus cyclicity, ovarian follicle numbers and maturation stage and reproductive organ weights; histopathology of reproductive organs (ovaries, uterus, cervix, vagina) is only performed when necessary or if organs have not been histologically examined in other multiple‐dose studies. b Number and maturation of ovarian follicles and number of corpora lutea in ovaries are not assessed in the OECD TG 421 (2016) and OECD TG 422 (2016). c Histopathology is performed in female animals of the control and high‐dose groups. Examination in other dose groups is extended when changes are seen in the highest dose groups. d Measurement is optional. e Ovaries, uterus and vagina histopathology is performed when necessary (e.g. in case of macroscopic lesions). f Litter size can be used as a surrogate in case there are no other data on implantation sites and the litter size is decreased. g Dose‐range finding studies are used to provide information on dose/concentration selection for studies of longer duration. Overview of reproductive toxicity parameters considered in regulatory assessment of female reproductive function and fertility. 443 (2018) 416 (2001) 443 (2018) 416 (2001) 415 (1983) 421 (2016) 422 (2016) 443 (2018) 416 (2001) 421 (2016) b 422 (2016) b 443 (2018) 416 (2001) 421 (2016) d 422 (2016) d [407 (2008) d ] g 408 (2018) 443 (2018) 421 (2016) d 422 (2016) d 407 (2008) d 443 (2018) 416 (2001) 415 (1983) e 421 (2016) e 422 (2016) [407 (2008)] g 408 (2018) 443 (2018) 421 (2016) e 422 (2016) [407 (2008)] g 408 (2018) 443 (2018) 416 (2001) 415 (1983) 421 (2016) 422 (2016) Abbreviations: EAS, Oestrogen, Androgen, Steroidogenesis endpoints described in the ECHA/EFSA Guidance on the identification of endocrine disruptors in the context of Regulations (EU) No 528/2012 and (EC) No 1107/2009 (2018). Older pesticide dossiers may only include one two‐generation reproductive toxicity study conducted according to the OECD TG 416 from 1983 and several parameters are not investigated, i.e. oestrus cyclicity, ovarian follicle numbers and maturation stage and reproductive organ weights; histopathology of reproductive organs (ovaries, uterus, cervix, vagina) is only performed when necessary or if organs have not been histologically examined in other multiple‐dose studies. Number and maturation of ovarian follicles and number of corpora lutea in ovaries are not assessed in the OECD TG 421 (2016) and OECD TG 422 (2016). Histopathology is performed in female animals of the control and high‐dose groups. Examination in other dose groups is extended when changes are seen in the highest dose groups. Measurement is optional. Ovaries, uterus and vagina histopathology is performed when necessary (e.g. in case of macroscopic lesions). Litter size can be used as a surrogate in case there are no other data on implantation sites and the litter size is decreased. Dose‐range finding studies are used to provide information on dose/concentration selection for studies of longer duration. In regulatory assessments, vaginal smears are commonly taken at sacrifice to correlate the stage of cycling with organs histology. In addition, evaluation of daily vaginal cytology samples is a simple and effective way to monitor the hypothalamic–hypophyseal–ovarian axis in rodents (Goldman et al., 2007). Vaginal cytology is a critical and sensitive endpoint, its primary value lies in the longitudinal assessment of the oestrous cycle and confirmation that animals used in a study are cycling normally. The number of females paired, with evidence of mating, with evidence of pregnancy and delivering a viable litter is used in the calculation of reproductive toxicity indices that are considered as measures of reproductive performance. Female reproductive indices are calculated for all dose levels in P generation animals (and, also, F1 generation animals depending on the study design), as indicated in the OECD Guidance Document No. 43 on mammalian reproductive toxicity testing and assessment (2008): TABLE 7Calculation of female reproductive indices.* Female reproductive indicesCalculation (%)DefinitionFemale mating index No.of sperm−positive femalesTotalNo.of females cohabitedx100 Measure of female's ability to mateFemale fertility index No.of pregnant femalesNo.of sperm−positive femalesx100 Measure of female's ability to become pregnantGestation index No.of females with live born pupsNo.of pregnant femalesx100 Measure of pregnancy that provides at least one live pup*Information extracted from Table 7 of the OECD GD No. 43 (2008). Calculation of female reproductive indices.* No.of sperm−positive femalesTotalNo.of females cohabitedx100 No.of pregnant femalesNo.of sperm−positive femalesx100 Information extracted from Table 7 of the OECD GD No. 43 (2008). Pre‐implantation losses, i.e. the number of embryos lost prior to implantation, may also be estimated through comparison of the numbers of corpora lutea and implantation sites for each dam and facilitate interpretation of fertility and gestation indices: Pre−implantation loss%=No.of corpora lutea−No.of implantation sitesNo.of corpora lutea×100 In a multi‐generation study, increased pre‐implantation loss could result from an adverse effect on gamete transport or function, resulting in failed fertilisation, or it can be a consequence of direct effects on the preimplantation embryo or indirect effects on the uterus or endocrine status of the dam (OECD GD No. 43). Consequently, besides impaired female fertility, an increase in pre‐implantation loss may also reflect an alteration of male gametes. Measurements of female reproductive hormone levels are not foreseen in currently available OECD test guidelines for reproductive toxicity (Table 2) and repeated dose short‐term toxicity (OECD TG 407, 408, 409). Substances that affect oestradiol production may be detected in the in vitro steroidogenesis assay (OECD TG 456, 2023) and in the in vitro aromatase assay (OPPTS 890.1200, 2011). Additional information on a potential (anti)oestrogenic effect of a substance suggesting disruption of the female reproductive system through an endocrine mode of action, may be generated in the in vitro OECD TG 455 (2021) oestrogen receptor (ER) transactivation assay, the ToxCast ER Bioactivity Model (US EPA CompTox Chemicals Dashboard 4 ) and the in vivo OECD TG 440 (2007) Uterotrophic bioassay in rodents. These in vitro and in vivo assays and tools (Table 8) that provide data on selected endocrine mechanism(s)/pathway(s) are described in the EFSA/ECHA Guidance on the identification of endocrine disruptors in the context of Regulations (EU) No 528/2012 and (EC) No 1107/2009 (2018). TABLE 8Recommended assays for the detection of oestrogen receptor (ant)agonists and substances affecting oestradiol production.Assay familyOECD test guideline (last update)Type of assayAssay objectiveTransactivation assayOECD TG 455 (2021) a Oestrogen receptor transactivation (ERTA) assayDetects agonists and antagonists of oestrogen receptor in vitro.OECD TG 440 (2007)Uterotrophic bioassay in rats: A short‐term screening assay for oestrogenic propertiesEvaluates the ability of a chemical to activate or inhibit the oestrogen receptor following ligand binding.SteroidogenesisOECD TG 456 (2023) b H295R steroidogenesis assayDetects substances that affect oestradiol production.Not available c Aromatase assayIdentifies chemicals that inhibit the catalytic activity of the cytochrome P450 enzyme aromatase (CYP19). Note: Other validated Guideline. a OPPTS 890.1300 (2009). b OPPTS 890.1550 (2011). c OPPTS 890.1200 (2011). Recommended assays for the detection of oestrogen receptor (ant)agonists and substances affecting oestradiol production. Note: Other validated Guideline. OPPTS 890.1300 (2009). OPPTS 890.1550 (2011). OPPTS 890.1200 (2011). In the TG 421, 422, 415, 416 and 443 studies, both sexes are treated, and therefore, it is usually not possible to determine whether an effect on reproductive performance is due to the female, the male or both without considering the other affected reproductive endpoints. In the same studies, the observed effects may in theory result from the exposure in utero, or from in‐life exposure, at any time of the life. Therefore, unless it can be concluded from the affected endpoints and/or generation that the effects result for the exposure during a well‐defined exposure window, the cumulative exposure assessments will need to cover consumer populations of all ages. In the case that stratification could be done, this will be considered during the uncertainty analysis and supported by appropriate sensitivity analysis. Challenges in regulatory tools investigating reproductive performance in regulatory assessments A weight‐of‐evidence approach (WoE) is followed in the assessment of all available data to determine whether there is a pattern of adversity on sexual function and fertility. Reproductive toxicity studies are only performed in rodents, the rat being the preferred species in respective guideline studies. For animal welfare purposes and due to the large number of animals used in these studies, the EU legislation prohibits their repetition, unless a clear justification indicates that this is necessary (e.g. data available are not reliable and/or not sufficient to conclude). Inconsistency in the findings among available guideline studies or in comparison to findings reported in studies published in the open literature adds to the uncertainty which should be described and quantified, if possible. Deviations from the OECD TG requirements, such as missing measurements on some reproductive toxicity endpoints or poor reporting of the methodology and/or results, are additional factors to be considered in the uncertainty analysis and assessment of study reliability and adequacy. For instance, for some active substances, only a two‐generation reproductive toxicity study conducted according to a former version of the OECD TG 416, i.e. version from 1983, is available. Therefore, for these substances, some parameters relevant for detecting adverse effects on sexual function and fertility were not investigated, e.g. sperm parameters, oestrus cyclicity and follicle counts. More information on differences in the analysis of parameters between OECD 416 versions are explained in detail in Table 3 (males) and 6 (females). More weight is given to studies with higher reliability when drawing the overall conclusion. Information on non‐rodent species may be obtained from the 90‐day (OECD TG 409, 1998) or 1‐year (OECD TG 452, 2018) repeated dose toxicity studies in the dog. However, it is often difficult to interpret the toxicological relevance of reproductive effects in the dog. In the 90‐day dog study, dosing could begin when animals are as young as 4–6 months, and therefore, at study termination, sexual maturity may not have been achieved. In pubertal dogs, spermatogenesis is relatively inefficient and not all tubules mature at the same time. As a result, the pubertal testes may contain few germ cells having undergone complete spermatogenesis but also some degenerating cells and the epididymes may only have few or no sperm in the lumen and often increased number of giant cells. Upon histological examination, it may be indistinguishable if the observed alterations (e.g. testicular atrophy, aspermia or low number of sperm count, giant spermatids) are due to sexually immature animals or to toxicity. In the female dog, vaginal cytology can be used to track the oestrous cycle, but the long duration and variability of the canine cycle makes this impractical in most nonclinical toxicology settings. Furthermore, reliable comparison of mean values of effects in reproductive parameters in treated dogs to the concurrent control mean values is hampered by the poor statistical power due to the low number of animals used in the study (typically 4 dogs/sex/group). For the 1‐year dog studies, similar considerations apply and confounding factors due to the long‐term treatment during the peripubertal period would still be present. Additionally, the 1‐year dog study is not a data requirement anymore. As a result of all these limitations in retrieving reliable reproductive toxicity information, studies in dogs are excluded from the parameters of male and female reproductive function and fertility described under Sections 2.3.2 and 2.3.3 and from data collection for the establishment of respective CAGs (see also Section 4.2.6). The identification and characterisation of general systemic toxicity as also described in the revised OECD GD No. 150 (2018) (e.g. signs of animal stress, mortality, changes in body weight and food consumption) is critical in deciding whether a substance exerts direct effect(s) on the reproductive system at doses below those eliciting systemic toxicity, or the observed adverse reproductive effects are only evidenced at doses where overt systemic toxicity is observed, and thus, they may be considered secondary, non‐specific consequences of the poor animal condition and loss of regenerative capacity. The criteria used by ECHA (2024) to account for excessive systemic toxicity for classifying substances for reproductive toxicity are considered in this project as a pragmatic approach to conclude on the inclusion of a substance into a CAG (see Section 4.3). In the context of endocrine disruption (ED) assessment, when the pattern of reproductive effects is deemed adverse, further investigation on the MoA is performed to conclude on the biological plausibility that the adverse effects are (exclusively) caused via an endocrine‐mediated MoA. Biological plausibility is stronger for the ‘EATS‐mediated’ parameters (including sperm motility/morphology/number, oestrus cyclicity, vaginal smears, reproductive organ weight and histopathology) in comparison to effects considered ‘sensitive to, but not diagnostic of, EATS parameters’ (including fertility, time to mating, conception, gestation length, dystocia, number of corpora lutea and implantations, number and maturation of ovarian follicles) (ECHA/EFSA guidance). In case there is no conclusive evidence on biological plausibility, or other MoA cannot be excluded, further empirical data need to be generated (see Section 3.1 on MoA/AOP) on the substance under evaluation to demonstrate the MoA and assist the assessment of toxicological significance and human relevance of the observed effect(s) (see also Section 2.4). MoA data may also be needed to assist interpretation of adverse effects on sexual function and fertility that are only observed in offspring when in adulthood. In case similar findings are not observed in parental generation animals, developmental origin cannot be excluded. As previously stated, currently available OECD test guidelines on reproductive toxicity studies (see Table 2) and repeated dose short‐term toxicity studies (OECD TG 407, 408, 409) do not include measurements of male and female reproductive hormones and/or other mechanistic information, which are only available on a case‐by‐case basis (e.g. in the context of ED assessments or to assess human relevance). Detailed testicular histopathology examination (e.g. using Bouin's fixative, paraffin embedding and transverse sections of 4–5 μm thickness) is conducted on the P and F1 males (cohort 1A and if needed, cohort 1B) in the EOGRT study (OECD TG 443, 2018) and the revised OECD TG 416 (2001), as well as the revised 90‐day oral toxicity study in rodents (OECD TG 408, 2018), in order to identify treatment‐related effects on testis differentiation and development, and on spermatogenesis, including retained spermatids, missing germ cell layers or types, multinucleated giant cells or sloughing of spermatogenic cells into the lumen (OECD GD No. 151, 2013). Details on the performance of fixation and histological examination are included in the OECD GD No. 106 (2009). In addition, detailed histological examination of the testes is essential in study protocols where the pre‐mating dosing period in males is limited (e.g. OECD TG 421, OECD TG 422), and therefore, fertility may not be a particularly sensitive indicator of testicular toxicity. Although detailed histological examination of testes is foreseen in several OECD test guidelines, the information is often missing from pesticide risk assessment dossiers, either because older versions of the study protocols were used (e.g. OECD TG 416 from 1983, OECD TG 408 from 1998) or because the study protocol is not part of the core data requirements (e.g. OECD TG 421, OECD TG 422). Overall, the data that are present in the RAR are expected to reflect the standard at the time when the study was performed. Detailed histopathological examination of the ovaries covering the follicular, luteal and interstitial compartments of the ovary, as well as the epithelial capsule and ovarian stroma (OECD GD No. 43, 2008) is not foreseen in the available OECD TGs and may be provided only when ovaries are target organs. Interpretation of reproductive findings in regulatory toxicological studies and their relevance to humans The ultimate aim of regulatory toxicological studies is the determination of potential hazard to human reproduction. Although alternative test methods can be expected to play an increasing role in the assessment of chemicals' reproductive toxicity, only animal models with hypothalamic–pituitary–gonadal axis integrity are intrinsically capable of capturing the complex biological processes involved in reproduction and account for absorption, distribution, metabolism and excretion of the tested chemicals. Reproductive processes are highly conserved through mammalian species and the rat has been the species of choice for multigenerational testing studies for decades. In respect to reproductive physiology similarities between humans and rats, effects on fertility in rodents seem to be a good indicator for effects in humans and by default the relevance to humans should be assumed, unless the opposite has been demonstrated (OECD GD No. 43, 2008). However, differences between rat and human in anatomy, physiology and timescales in normal biological processes must be taken into consideration when translating reproductive findings in rat to the human condition. A major difference to be considered is that humans are of relatively low fertility and thus may be at greater risk from reproductive toxicants than rats. The use of multiple endpoints in reproductive toxicity studies and the standard uncertainty factor of 10 for interspecies variability are expected to cover for such differences. Predictivity of the rat model for men fertility As reported by Mangelsdorf et al. (2003), male rats and rabbits remain fertile even when their sperm count is approximately 10% and 1% of control levels, respectively. In humans, sperm concentration is lower than in rodents and even a small decrease would lead to fertility issues. Indeed, fertility is considerably reduced in men with sperm count being 30% of the physiological level. Male rats have a large excess of spermatozoa in their ejaculates. Therefore, effect on fertility index alone in a rat reproductive toxicity study is not a sensitive indicator to predict fertility in humans. As such, additional parameters should be included in the overall assessment of male fertility. A highly sensitive endpoint is histopathology of the testes, preferably using a refined method describing different stages of spermatogenesis. Other sensitive endpoints are sperm parameters such as sperm count, sperm morphology and sperm motility. The weights of reproductive organs including accessory glands, i.e. testis, epididymis, prostate, seminal vesicles provide useful additional information of confirmatory nature. Regarding sperm parameters, sperm motility was found to be in some cases more sensitive than histopathology because direct effect on the epididymis or the sperm can affect motility (OECD GD No. 43, 2008). Also, changes in sperm morphology and sperm counts should not be discarded in the absence of histological lesions (OECD GD No. 43, 2008). Overall, changes in histopathology and/or sperm parameters should be considered as indicative of a potential effect on fertility in humans even if fertility in rats is not affected. Predictivity of the rat model for women fertility While for both rats and humans the primordial follicle stock serves as a reserve for the entire reproductive life span of the adult, menopause in women is the consequence of follicular exhaustion within the ovary, while in rats reproductive senescence results from initial centrally mediated changes with alterations in hypothalamic function (Finch, 2014; Gore et al., 2000; Vidal, 2017). In reproductive studies in rats, a change in follicle number often precedes a change in organ weight or histopathology. The magnitude of the reduction in the number of follicles will determine whether there is an effect on reproductive performance (OECD GD No. 43, 2008). In respect to the particular concern of a depletion of follicle reserves in women (premature ovarian insufficiency, premature menopause), decreased follicle counts observed in rat studies should be considered as indicative of a potential effect on fertility in women even if no other reproductive endpoint is affected. ADME considerations for the predictivity of animal models to human fertility Although possible species‐specific differences in pharmacokinetics parameters (including metabolism) may occur, adverse effects on reproductive function observed in rats are assumed to be predictive of the same effect in humans in the absence of data proving the contrary. For active substances, no data on human ADME parameters are generally available. A comparative in vitro metabolism study, conducted on animal and human test systems, is required for each active substance. The results of this study may be used to determine the human relevance of adverse effects observed in animal species in a weight of evidence approach. However, a species‐specific effect can only be tentatively attributed to the formation of a ‘unique’ metabolite and further mechanistic studies on the plausibility of the hypothesis are needed (EFSA PPR Panel, 2021). DATA AND METHODOLOGIES Methodology for hazard identification: Specific effects and indicators of reproductive function Definitions of ‘specific effect’, ‘indicator’ and ‘ancillary finding’ In the context of this report, a specific effect is defined as the precise toxicological insult that can result from a combined action of pesticides and is relevant for grouping pesticides into CAGs for CRA (EFSA PPR Panel, 2013). The criteria set by the EFSA PPR Panel (2013), i.e. adversity, systemic and unambiguous nature of the effect (primary effect, MoAs/Adverse Outcome Pathways (AOPs)) and relevance to humans, are applied by the WG by expert judgement to identify specific effects on reproductive function based on relevant toxicity endpoint(s), i.e. indicator(s). It should be noted that ‘specific effects’ are not necessarily described by a single toxicity endpoint but rather by a set of endpoints that make up an effect of greater biological complexity, e.g. decreased fertility. An indicator of a specific effect is a toxicological endpoint identifiable in toxicological studies reflecting the specific effect and useful for characterising the toxicological potency of the pesticide causing that effect. A single or a set of indicators can be used to identify a specific effect. These parameters will be collected for all substances. An ancillary finding is a change in a male/female reproductive parameter that in isolation does not trigger the direct inclusion of an active substance into a CAG. However, since it can support the indicator(s), it can increase the likelihood that an active substance has been correctly allocated to a CAG. Ancillary findings will only be collected for risk drivers (a.s. identified on the basis of hazard and exposure considerations among the list of prioritised substances) in a second step of the data collection as additional evidence that supports CAG membership probability. A risk driver is defined as a pesticide/commodity combination contributing, in Tier II of the cumulative exposure calculation, at least 5% of the cumulative exposures exceeding the 99th percentile estimate in at least one of the assessed populations. For detailed information on the methodology used to identify risk drivers, see EFSA (2022). Hazard‐driven criteria for grouping pesticides into cumulative assessment groups (CAGs) Among all possible effects of chemicals on the reproductive function including fertility, those relevant for grouping pesticides into CAGs were identified using hazard‐driven criteria as advised by the guidance of the EFSA Scientific Committee on scientific criteria for grouping chemicals into assessment groups for human risk assessment of combined exposure to multiple chemicals (EFSA Scientific Committee, 2021). The framework proposed by the EFSA Scientific Committee is based on the use of a hierarchical approach to evaluate the body of information available for each a.s., as described by the picture below (Figure 5): Top‐down hierarchical process for grouping chemicals into assessment groups using hazard‐driven criteria (Source: EFSA Scientific Committee, 2021). In case of incomplete mechanistic information, the grouping may be based on the commonality of adverse outcomes. This approach is also of interest in case of AOP networks converging to the same adverse outcome even if they have different initiating events and pathways. The grouping based on common target organ/system toxicity implies a large uncertainty when the dose‐addition model is used for quantifying the combined effects of groups of chemicals assembled in this way. Using common MoA/AOP – relevance of AOPs to identify reproductive specific effects According to the EFSA Scientific Committee's framework, the golden standard for grouping chemicals is considered to be the common mode of action (MoA) and/or the adverse outcome pathways (AOPs). For the identification of AOPs on reproductive specific effects, the terms ‘reproduction’, ‘reproductive’, ‘fertility’, ‘fecundity’ and ‘sperm’ were used as keywords for searching within the title or full text in the Collaborative Adverse Outcome Pathway Wiki (AOPwiki) as of April 7, 2024 (AOPwiki, Version 2.7, released on 30 March 2024). The list of retrieved AOPs was cross‐checked for completeness with the list of AOPs, key events (KEs), molecular initiating events (MIEs) and adverse outcomes (AOs) that are listed in the AOPwiki supporting file (https://aopwiki.org/downloads/aop_ke_mie_ao.tsv) and additional relevant AOPs were retrieved (see Annex A). The following criteria were applied for the selection of relevant AOPs in the context of this project: Chemically induced AOs. When the MIE included ionising radiation (MIE1686: Deposition of energy, MIE1550: Deposition of ionising energy) the AOP was not considered relevant for this project and it was excluded from the list of retrieved AOPs. Human relevance. The retrieved AOPs were considered of (potential) human relevance when there was information on taxonomic applicability that included humans and/or other mammal(s) or in case information on taxonomy was lacking, human relevance could not be excluded based on current scientific knowledge and relevant information from the title, KEs or AOs (e.g. AOP126 alpha‐noradrenergic antagonism leads to reduced fecundity via delayed ovulation, AOP73 xenobiotic inhibition of dopamine‐beta‐hydroxylase and subsequent reduced fecundity, etc.). Overrepresentation of ecotoxicological AOs was observed, and according to recent scientific review (Jaylet et al., 2024), it is attributed to the fact that the AOP concept was initially established to address ecotoxicological questions (Ankley et al., 2010). Human relevance of available AOPs may be revised in the future based on the outcome of ongoing research on weight of evidence for cross‐species conservation of androgen receptor‐based biological activity (Vliet et al., 2023) and on advancement of cross‐species extrapolation of the effects of chemicals in regulatory toxicology (Lalone et al., 2021). Non‐cancerous AOs. The AOP440 (hypothalamus oestrogen receptors activity suppression leading to ovarian cancer via ovarian epithelial cell hyperplasia) where the AOs were promotion of ovarian adenomas (AO1053) and promotion of ovarian cancer (AO2092) was excluded from the list of relevant AOPs for the assessment of reproductive toxicity effects. Chemically induced AOs. When the MIE included ionising radiation (MIE1686: Deposition of energy, MIE1550: Deposition of ionising energy) the AOP was not considered relevant for this project and it was excluded from the list of retrieved AOPs. Chemically induced AOs. When the MIE included ionising radiation (MIE1686: Deposition of energy, MIE1550: Deposition of ionising energy) the AOP was not considered relevant for this project and it was excluded from the list of retrieved AOPs. Human relevance. The retrieved AOPs were considered of (potential) human relevance when there was information on taxonomic applicability that included humans and/or other mammal(s) or in case information on taxonomy was lacking, human relevance could not be excluded based on current scientific knowledge and relevant information from the title, KEs or AOs (e.g. AOP126 alpha‐noradrenergic antagonism leads to reduced fecundity via delayed ovulation, AOP73 xenobiotic inhibition of dopamine‐beta‐hydroxylase and subsequent reduced fecundity, etc.). Overrepresentation of ecotoxicological AOs was observed, and according to recent scientific review (Jaylet et al., 2024), it is attributed to the fact that the AOP concept was initially established to address ecotoxicological questions (Ankley et al., 2010). Human relevance of available AOPs may be revised in the future based on the outcome of ongoing research on weight of evidence for cross‐species conservation of androgen receptor‐based biological activity (Vliet et al., 2023) and on advancement of cross‐species extrapolation of the effects of chemicals in regulatory toxicology (Lalone et al., 2021). Human relevance. The retrieved AOPs were considered of (potential) human relevance when there was information on taxonomic applicability that included humans and/or other mammal(s) or in case information on taxonomy was lacking, human relevance could not be excluded based on current scientific knowledge and relevant information from the title, KEs or AOs (e.g. AOP126 alpha‐noradrenergic antagonism leads to reduced fecundity via delayed ovulation, AOP73 xenobiotic inhibition of dopamine‐beta‐hydroxylase and subsequent reduced fecundity, etc.). Overrepresentation of ecotoxicological AOs was observed, and according to recent scientific review (Jaylet et al., 2024), it is attributed to the fact that the AOP concept was initially established to address ecotoxicological questions (Ankley et al., 2010). Human relevance of available AOPs may be revised in the future based on the outcome of ongoing research on weight of evidence for cross‐species conservation of androgen receptor‐based biological activity (Vliet et al., 2023) and on advancement of cross‐species extrapolation of the effects of chemicals in regulatory toxicology (Lalone et al., 2021). Non‐cancerous AOs. The AOP440 (hypothalamus oestrogen receptors activity suppression leading to ovarian cancer via ovarian epithelial cell hyperplasia) where the AOs were promotion of ovarian adenomas (AO1053) and promotion of ovarian cancer (AO2092) was excluded from the list of relevant AOPs for the assessment of reproductive toxicity effects. Non‐cancerous AOs. The AOP440 (hypothalamus oestrogen receptors activity suppression leading to ovarian cancer via ovarian epithelial cell hyperplasia) where the AOs were promotion of ovarian adenomas (AO1053) and promotion of ovarian cancer (AO2092) was excluded from the list of relevant AOPs for the assessment of reproductive toxicity effects. Outcome of the available AOP analysis. Overall, 32 relevant AOPs were identified for reproductive specific effects from AOPwiki, (Version 2.7, released on 30 March 2024). Of these, 21 AOPs were applicable to males, 10 AOPs were applicable to females and AOP492 was reported as unspecific and was considered applicable to both males and females. The list of relevant AOPs is collected in a separate Excel file (Annex A.). Male reproduction From the retrieved AOPs on male reproductive specific effects, only AOP212 (histone deacetylase inhibition leading to testicular atrophy) is already endorsed by the OECD (Figure 6). AOP212‐Histone deacetylase inhibition leading to testicular atrophy, endorsed by the OECD (Source: https://aopwiki.org/aops/212). The AOP18 (PPARα activation in utero leading to impaired fertility in males) is currently under review by the Extended Advisory Group for Molecular Screening and Toxicogenomics (EAGMST) and AOPs 51 (PPARα activation leading to impaired fertility in adult male rodents), 64 (Glucocorticoid Receptor (GR) Mediated Adult Leydig Cell Dysfunction Leading to Decreased Male Fertility (Adult Leydig Cell Dysfunction)), 305 (5α‐reductase inhibition leading to short anogenital distance (AGD) in male (mammalian) offspring), 306 (Androgen receptor (AR) antagonism leading to short anogenital distance (AGD) in male (mammalian) offspring) and 307 (Decreased testosterone synthesis leading to short anogenital distance (AGD) in male (mammalian) offspring) are under development. For all other AOPs, the OECD status is not publicly available. The AOs in the identified AOPs for specific effects on male reproductive system are the following: Decrease, fertility (AO330)Impairment of reproductive capacity (AO337)Impaired fertility (AO406) ○Malformation, male reproductive tract (AO348) Decreased sperm quantity/quality in the adult, decreased fertility (AO505)Decreased sperm quantity or quality in the adult, decreased fertility (AO520)Decreased fertility in the adult, decreased sperm quantity and/or quality in the adult test (AO543)Decreased cholesterol, decreased sperm quantity and/or quality in the adult testis (AO646)Testicular atrophy (AO1506)Reduce, sperm count (AO1757)Impaired, spermatogenesis (AO1758)Decrease, reproduction (AO1863) ○Decreased sperm quantity or quality in the adult, decreased fertility (AO520)○Malformation, cryptorchidism – maldescended testis (AO1616)○Anogenital distance (AGD), decreased (AO1688) Decreased, viable offspring (AO2147) Decrease, fertility (AO330) Decrease, fertility (AO330) Impairment of reproductive capacity (AO337) Impairment of reproductive capacity (AO337) Impaired fertility (AO406) ○Malformation, male reproductive tract (AO348) Impaired fertility (AO406) ○Malformation, male reproductive tract (AO348) ○Malformation, male reproductive tract (AO348) Malformation, male reproductive tract (AO348) Decreased sperm quantity/quality in the adult, decreased fertility (AO505) Decreased sperm quantity/quality in the adult, decreased fertility (AO505) Decreased sperm quantity or quality in the adult, decreased fertility (AO520) Decreased sperm quantity or quality in the adult, decreased fertility (AO520) Decreased fertility in the adult, decreased sperm quantity and/or quality in the adult test (AO543) Decreased fertility in the adult, decreased sperm quantity and/or quality in the adult test (AO543) Decreased cholesterol, decreased sperm quantity and/or quality in the adult testis (AO646) Decreased cholesterol, decreased sperm quantity and/or quality in the adult testis (AO646) Testicular atrophy (AO1506) Testicular atrophy (AO1506) Reduce, sperm count (AO1757) Reduce, sperm count (AO1757) Impaired, spermatogenesis (AO1758) Impaired, spermatogenesis (AO1758) Decrease, reproduction (AO1863) ○Decreased sperm quantity or quality in the adult, decreased fertility (AO520)○Malformation, cryptorchidism – maldescended testis (AO1616)○Anogenital distance (AGD), decreased (AO1688) Decrease, reproduction (AO1863) ○Decreased sperm quantity or quality in the adult, decreased fertility (AO520)○Malformation, cryptorchidism – maldescended testis (AO1616)○Anogenital distance (AGD), decreased (AO1688) ○Decreased sperm quantity or quality in the adult, decreased fertility (AO520) Decreased sperm quantity or quality in the adult, decreased fertility (AO520) ○Malformation, cryptorchidism – maldescended testis (AO1616) Malformation, cryptorchidism – maldescended testis (AO1616) ○Anogenital distance (AGD), decreased (AO1688) Anogenital distance (AGD), decreased (AO1688) Decreased, viable offspring (AO2147) Decreased, viable offspring (AO2147) It is noted that some AOPs have more than one adverse outcome. Decreased reproduction (AO1863) was encountered as a consequence of decreased sperm quantity or quality and decreased fertility (AO520), cryptorchidism (AO1616) or decreased anogenital distance (AO1688). Impaired fertility (AO406) was the only adverse outcome in AOP51 and AOP64 and it also appeared as a consequence of male reproductive tract malformations (AO348). Decreased sperm quantity or quality and decreased fertility appeared both as an adverse outcome (AO520) leading to decreased reproduction (AO1863) in the AOP476, as the only adverse outcome in AOP70 and AOP71 and as a key event in AOP64 leading to impaired fertility (AO406). Decreased anogenital distance (AO1688) was also reported as the sole adverse outcome in three additional AOPs (AOP305, AOP306, AOP307). Although these AOPs are not currently validated and a link to reproductive dysfunction is not part of the AOPs, decreased anogenital distance has been reported as a toxicological or clinical marker for fetal androgen action and risk for reproductive disorders (Schwartz et al., 2019) and it is therefore considered to be a relevant adverse outcome for this project. Three broad modes of action areas were identified for specific effects on the male reproductive system, mostly on the basis of the MIEs (supported by KEs) included in the AOPs, i.e. endocrine disruption, genotoxicity, oxidative stress: Endocrine disruption , through androgen receptor antagonism (MIE26, MIE27), activation of PPARα (MIE227), decreased testosterone synthesis in Leydig cells (MIE413), decreased testosterone by the fetal Leydig cells (MIE653, MIE654, MIE658, MIE659, MIE661, MIE662), glucocorticoid receptor agonism (MIE494), increased glucocorticoid (MIE650), decreased steroidogenesis (MIE640, MIE804, MIE1617, KE647), decreased cholesterol through decreased biosynthesis (MIE642), uptake of lipoproteins (MIE643) and transport to inner mitochondrial membrane (MIE644) leading to decreased testosterone production by adult Leydig Cells (KE645) and disruption of the retinoic acid pathway through MIE1888 (Increased (ectopic) all‐trans retinoic acid concentration). Genotoxicity , including histone deacetylase inhibition (ΜΙΕ1502) leading to spermatocyte depletion (KE1515) and testicular atrophy (AO1506) or DNA alkylation (ΜΙΕ322) leading to increased apoptosis (KE1365) and reduced sperm count (AO1757). Oxidative stress , through increased production of reactive oxygen species (ΜΙΕ1115 or KE1115) leading to e.g. apoptosis (KE1262, KE496) or lipid peroxidation (KE1445) as intermediate key events. It is acknowledged that oxidative stress is not specific to reproductive effects and thus it cannot be considered as a sole predictor of reproductive toxicity. In addition, measurements of oxidative stress are not included in validated OECD test guidelines and such information is not usually available in regulatory toxicology studies. Endocrine disruption , through androgen receptor antagonism (MIE26, MIE27), activation of PPARα (MIE227), decreased testosterone synthesis in Leydig cells (MIE413), decreased testosterone by the fetal Leydig cells (MIE653, MIE654, MIE658, MIE659, MIE661, MIE662), glucocorticoid receptor agonism (MIE494), increased glucocorticoid (MIE650), decreased steroidogenesis (MIE640, MIE804, MIE1617, KE647), decreased cholesterol through decreased biosynthesis (MIE642), uptake of lipoproteins (MIE643) and transport to inner mitochondrial membrane (MIE644) leading to decreased testosterone production by adult Leydig Cells (KE645) and disruption of the retinoic acid pathway through MIE1888 (Increased (ectopic) all‐trans retinoic acid concentration). Endocrine disruption , through androgen receptor antagonism (MIE26, MIE27), activation of PPARα (MIE227), decreased testosterone synthesis in Leydig cells (MIE413), decreased testosterone by the fetal Leydig cells (MIE653, MIE654, MIE658, MIE659, MIE661, MIE662), glucocorticoid receptor agonism (MIE494), increased glucocorticoid (MIE650), decreased steroidogenesis (MIE640, MIE804, MIE1617, KE647), decreased cholesterol through decreased biosynthesis (MIE642), uptake of lipoproteins (MIE643) and transport to inner mitochondrial membrane (MIE644) leading to decreased testosterone production by adult Leydig Cells (KE645) and disruption of the retinoic acid pathway through MIE1888 (Increased (ectopic) all‐trans retinoic acid concentration). Genotoxicity , including histone deacetylase inhibition (ΜΙΕ1502) leading to spermatocyte depletion (KE1515) and testicular atrophy (AO1506) or DNA alkylation (ΜΙΕ322) leading to increased apoptosis (KE1365) and reduced sperm count (AO1757). Genotoxicity , including histone deacetylase inhibition (ΜΙΕ1502) leading to spermatocyte depletion (KE1515) and testicular atrophy (AO1506) or DNA alkylation (ΜΙΕ322) leading to increased apoptosis (KE1365) and reduced sperm count (AO1757). Oxidative stress , through increased production of reactive oxygen species (ΜΙΕ1115 or KE1115) leading to e.g. apoptosis (KE1262, KE496) or lipid peroxidation (KE1445) as intermediate key events. It is acknowledged that oxidative stress is not specific to reproductive effects and thus it cannot be considered as a sole predictor of reproductive toxicity. In addition, measurements of oxidative stress are not included in validated OECD test guidelines and such information is not usually available in regulatory toxicology studies. Oxidative stress , through increased production of reactive oxygen species (ΜΙΕ1115 or KE1115) leading to e.g. apoptosis (KE1262, KE496) or lipid peroxidation (KE1445) as intermediate key events. It is acknowledged that oxidative stress is not specific to reproductive effects and thus it cannot be considered as a sole predictor of reproductive toxicity. In addition, measurements of oxidative stress are not included in validated OECD test guidelines and such information is not usually available in regulatory toxicology studies. Female reproduction None of the retrieved AOPs on female reproductive specific effects is endorsed by the OECD. The AOP7 (Aromatase (Cyp19a1) reduction leading to impaired fertility in adult female) is currently under review by the EAGMST and the AOP345 (Androgen receptor (AR) antagonism leading to decreased fertility in females) and AOP398 (Inhibition of ALDH1A (RALDH) leading to impaired fertility via disrupted meiotic initiation of fetal oogonia of the ovary) are under development. For all other AOPs the OECD status is not available. The AOs in the identified AOPs for specific effects on the female reproductive system are the following: Irregularities, ovarian cycle (AO405)Impaired, fertility (AO406)Reduced, reproductive success (AO675)Decrease, population growth rate (AO360)Decreased, decreased fecundity (AO527)Decreased fertility, reduced number of oocytes ovulated (AO972)Impaired, reproduction (AO1991) Irregularities, ovarian cycle (AO405) Irregularities, ovarian cycle (AO405) Impaired, fertility (AO406) Impaired, fertility (AO406) Reduced, reproductive success (AO675) Reduced, reproductive success (AO675) Decrease, population growth rate (AO360) Decrease, population growth rate (AO360) Decreased, decreased fecundity (AO527) Decreased, decreased fecundity (AO527) Decreased fertility, reduced number of oocytes ovulated (AO972) Decreased fertility, reduced number of oocytes ovulated (AO972) Impaired, reproduction (AO1991) Impaired, reproduction (AO1991) As with AOPs on male reproductive specific effects, in females, it is also noted that some AOPs have more than one adverse outcome. Impaired fertility (AO406) may result from irregularities in the ovarian cycle described by AO405 in AOP7 or KE405 in AOP345 and AOP398. Decreased reproductive success (AO675) may result in a decreased population growth rate (AO360). The same broad mode of action areas were identified for specific effects on the female reproductive system, as with males, mostly on the basis of the MIEs (supported by KEs), i.e. endocrine disruption and genotoxicity: Endocrine disruption , through androgen receptor antagonism (MIE26), activation of oestrogen receptor alpha (MIE1065), inhibition of aromatase (MIE408, MIE964), cyclooxygenase (MIE79, MIE102, MIE103), dopamine‐beta‐hydroxylase (MIE526), retinaldehyde dehydrogenase 1A (ALDH1A) (MIE1880) or antagonism of the alpha‐adrenergic receptor (NE receptor) in the hypothalamus (MIE849). Genotoxicity, evidenced as downstream KEs of cyclooxygenase inhibition leading to reproductive dysfunction via either interference with meiotic prophase I /metaphase I transition (MIE102, KE694 – Reduced Meiotic prophase I/metaphase I transition, oocyte) or interference with spindle assembly checkpoint (MIE103, KE695‐Upregulated, Spindle assembly checkpoint protein Mad2‐oocyte). It is however noted that these specific genotoxicity endpoints are not routinely investigated in regulatory genotoxicity data package. However, upstream KEs of MIE102 and MIE103 include changes in hormonal measurements that could be included in mechanistic testing of reproductive toxicity (e.g. KE690‐reduced luteinising hormone (LH) in plasma). Oxidative stress , through increased production of reactive oxygen species (KE1115) leading to lipid peroxidation (KE1445) as intermediate key events. It is acknowledged that oxidative stress is not specific to reproductive effects, and thus, it cannot be considered as a sole predictor of reproductive toxicity. In addition, measurements of oxidative stress are not included in validated OECD test guidelines and such information is not usually available in regulatory toxicology studies. Endocrine disruption , through androgen receptor antagonism (MIE26), activation of oestrogen receptor alpha (MIE1065), inhibition of aromatase (MIE408, MIE964), cyclooxygenase (MIE79, MIE102, MIE103), dopamine‐beta‐hydroxylase (MIE526), retinaldehyde dehydrogenase 1A (ALDH1A) (MIE1880) or antagonism of the alpha‐adrenergic receptor (NE receptor) in the hypothalamus (MIE849). Endocrine disruption , through androgen receptor antagonism (MIE26), activation of oestrogen receptor alpha (MIE1065), inhibition of aromatase (MIE408, MIE964), cyclooxygenase (MIE79, MIE102, MIE103), dopamine‐beta‐hydroxylase (MIE526), retinaldehyde dehydrogenase 1A (ALDH1A) (MIE1880) or antagonism of the alpha‐adrenergic receptor (NE receptor) in the hypothalamus (MIE849). Genotoxicity, evidenced as downstream KEs of cyclooxygenase inhibition leading to reproductive dysfunction via either interference with meiotic prophase I /metaphase I transition (MIE102, KE694 – Reduced Meiotic prophase I/metaphase I transition, oocyte) or interference with spindle assembly checkpoint (MIE103, KE695‐Upregulated, Spindle assembly checkpoint protein Mad2‐oocyte). It is however noted that these specific genotoxicity endpoints are not routinely investigated in regulatory genotoxicity data package. However, upstream KEs of MIE102 and MIE103 include changes in hormonal measurements that could be included in mechanistic testing of reproductive toxicity (e.g. KE690‐reduced luteinising hormone (LH) in plasma). Genotoxicity, evidenced as downstream KEs of cyclooxygenase inhibition leading to reproductive dysfunction via either interference with meiotic prophase I /metaphase I transition (MIE102, KE694 – Reduced Meiotic prophase I/metaphase I transition, oocyte) or interference with spindle assembly checkpoint (MIE103, KE695‐Upregulated, Spindle assembly checkpoint protein Mad2‐oocyte). It is however noted that these specific genotoxicity endpoints are not routinely investigated in regulatory genotoxicity data package. However, upstream KEs of MIE102 and MIE103 include changes in hormonal measurements that could be included in mechanistic testing of reproductive toxicity (e.g. KE690‐reduced luteinising hormone (LH) in plasma). Oxidative stress , through increased production of reactive oxygen species (KE1115) leading to lipid peroxidation (KE1445) as intermediate key events. It is acknowledged that oxidative stress is not specific to reproductive effects, and thus, it cannot be considered as a sole predictor of reproductive toxicity. In addition, measurements of oxidative stress are not included in validated OECD test guidelines and such information is not usually available in regulatory toxicology studies. Oxidative stress , through increased production of reactive oxygen species (KE1115) leading to lipid peroxidation (KE1445) as intermediate key events. It is acknowledged that oxidative stress is not specific to reproductive effects, and thus, it cannot be considered as a sole predictor of reproductive toxicity. In addition, measurements of oxidative stress are not included in validated OECD test guidelines and such information is not usually available in regulatory toxicology studies. Mechanistic information on endocrine disruption from ToxCast Several specific effects used to develop the CAGs for reproductive toxicity are also used for the assessment of the endocrine‐disrupting properties of pesticide active substances and can therefore benefit not only of the currently available or under development AOPs but also of the regulatory documents developed to help in their identification. The OECD GD 150 (2018) and the ECHA/EFSA ED GD (EFSA, 2018) indicate which of these specific effects should be considered as a likely consequence of the disruption of an endocrine pathway, based on current toxicological and endocrinological knowledge. In addition, they propose a methodology based on the AOP/MOA conceptual framework to contextualise the available evidence for postulating the most likely affected endocrine modality. Although this is restricted to the oestrogenic, androgenic and steroidogenesis (EAS) pathways, where feasible, allocation of substances can be refined based on a postulated MOA, even if the available information is only the specific effect. Additional substance‐specific mechanistic information can be retrieved from the U.S. EPA dashboard ToxCast programme which is offering high‐throughput screening data for the assessment of the EAS activities that can be used as an alternative for regulatory guideline studies. Several pesticides are included in the database and for several substances in addition to the high‐throughput screening data, computational models for the E and A modalities are also available. Along the regulatory peer review pesticides authorisation process, several active substances have been assessed for their endocrine‐disrupting properties following the ECHA/EFSA ED GD. This information can be used for the identification, characterisation and MOA analysis of specific effects affecting reproduction through an endocrine‐mediated MOA and this can also in a mechanistically based refinement of the groups, if necessary. Using common toxicological effects to identify specific effects Considering the limitations of the relevant AOPs mentioned above and the lack of information on the MoA leading to effects on reproductive function and fertility for several a.s. and metabolites, the WG considered appropriate to group the a.s. and metabolites in the respective CAGs based on the common adverse outcomes captured as apical toxicity endpoints in regulatory studies, when these can be reasonably assumed to result from a molecular initiating event (MIE) or/and KEs taking place in the reproductive system. The specific effects (i.e. toxicological effects, AOs) of pesticides on male and female reproductive function and fertility, and their indicators, were defined by the WG using scientific knowledge and expert judgement. Methodology for substance prioritisation To minimise the resources needed for the establishment of CAGs and to focus on the a.s. predominating in food, EFSA developed and implemented a prioritisation methodology (EFSA, 2024). In brief, this methodology is based on short‐ and long‐term exposure assessments for each individual pesticide quantified in at least one food sample of the official monitoring program of EU Member States over a 3‐year monitoring cycle from 2019 to 2021. These assessments were performed by probabilistic modelling in 30 population groups, covering different age groups (adults, children and toddlers) in 17 countries. The threshold to prioritise a pesticide was an acute or chronic hazard quotient (HQ), ratio between the acute or chronic exposure from monitoring studies to the acute reference dose (ARfD) or acceptable daily intake (ADI), respectively, equal to or exceeding 0.1 (10% of the reference value) at P99.9 of the exposure distribution in at least one of the 30 population groups under consideration. This resulted in a priority list of 67 substances, of which 21 reached the threshold in acute exposure assessments only, 15 in long‐term exposure assessments only and 31 in both acute and long‐term exposure assessments. The REPRO CRA will not cover effects due to acute exposure. The WG recognised that, mechanistically, acute exposure can influence the outcome of some specific effects. However, the WG is of the opinion that separation of acute vs. chronic exposure determining the effect is not practically feasible and only chronic exposure will be used based on the study designs. In addition, the WG recognised that setting of ARfD is generally not based on reproductive (i.e. fertility) endpoints. Only 46 substances will be considered from the list of prioritised substances based on long‐term exposure. The WG noted that the metabolites of these prioritised substances are not included in the monitoring studies, and therefore, they could be missed. Therefore, any metabolite of these 46 prioritised substances that is spotted in the available documentation and considered as a reproductive toxicant will be included in the data collection. EFSA will check if relevant reproductive toxicity data have in the meantime been submitted on these substances and whether these data have been peer reviewed (in an expert meeting). ASSESSMENT Hazard identification Specific effects and indicators of male reproductive function including fertility Two specific effects have been identified for male reproduction: 1. Altered mating performance and 2. Decreased fertility. The respective indicators of these specific effects are listed in Table 9. For the definitions and calculations of male reproductive indices as per OECD GD No. 43, please refer to Section 2.3.2 and Table 4. TABLE 9Specific effects and indicators of male reproductive function including fertility.Specific effectIndicatorAncillary finding (only for risk drivers)1. Altered mating performance Delayed time to matingDecreased male mating index 2. Decreased fertility Decreased male fertility indexDecreased number of implantation sites (or increased pre‐implantation loss)* Decreased litter size* Changes in sperm parameters (number, morphology, and/or motility)Change in testes histopathology (seminiferous epithelium and Leydig cells) Change in testis weightChange in Epididymis weightChange in epididymis histopathologyChange in prostate weightChange in prostate histopathologyChange in seminal vesicles (with coagulating gland) weightChange in seminal vesicles histopathologyChange in coagulating gland histopathologyLevator ani/bulbocavernosus muscle complex (LABC) weightChange in Cowper's gland weightChange in glans penis weight *While these parameters are measured in females, besides impairment of female fertility, they may also be indicative of a functional deficit related to sperm. Specific effects and indicators of male reproductive function including fertility. Delayed time to matingDecreased male mating index Delayed time to mating Delayed time to mating Decreased male mating index Decreased male mating index Decreased male fertility index Decreased male fertility index Decreased number of implantation sites (or increased pre‐implantation loss)* Decreased number of implantation sites (or increased pre‐implantation loss)* Decreased litter size* Decreased litter size* Changes in sperm parameters (number, morphology, and/or motility) Changes in sperm parameters (number, morphology, and/or motility) Change in testes histopathology (seminiferous epithelium and Leydig cells) Change in testes histopathology (seminiferous epithelium and Leydig cells) Change in testis weight Change in testis weight Change in Epididymis weight Change in Epididymis weight Change in epididymis histopathology Change in epididymis histopathology Change in prostate weight Change in prostate weight Change in prostate histopathology Change in prostate histopathology Change in seminal vesicles (with coagulating gland) weight Change in seminal vesicles (with coagulating gland) weight Change in seminal vesicles histopathology Change in seminal vesicles histopathology Change in coagulating gland histopathology Change in coagulating gland histopathology Levator ani/bulbocavernosus muscle complex (LABC) weight Levator ani/bulbocavernosus muscle complex (LABC) weight Change in Cowper's gland weight Change in Cowper's gland weight Change in glans penis weight Change in glans penis weight While these parameters are measured in females, besides impairment of female fertility, they may also be indicative of a functional deficit related to sperm. All findings will be collected and more detailed criteria for inclusion of a substance in a CAG are described in 4.3. So‐called ancillary findings, i.e. changes in male reproductive parameters in studies where at least one indicator has been observed that can increase the likelihood that a substance has been correctly allocated to a CAG are also listed in Table 9 and will be collected for risk drivers only. Changes in clinical chemistry parameters are not considered suitable indicators for identification of the specific effects for male reproductive function due to lack of data since measurements of male reproductive hormone levels are not foreseen in the currently available OECD test guidelines on reproductive and repeated dose short‐term toxicity (see also Section 2.3.2). However, when such measurements are available, they will be collected only for risk drivers to assess whether disruption of the male reproductive system occurs through an endocrine mode of action or not. Specific effects and indicators of female reproductive function including fertility Four specific effects were identified for female reproduction: (1) Altered mating performance, (2) decreased fertility, (3) altered gestation and (4) altered parturition. The respective indicators of these specific effects are listed in Table 10. For the definitions and calculations of female reproductive indices as per OECD GD No, 43, please refer to Section 2.3.3 and Table 7. TABLE 10Specific effects and indicators of female reproductive function including fertility.Specific effectIndicatorAncillary finding (only for risk drivers)1. Altered mating performance Delayed time to matingReduced female mating index 2. Decreased Fertility Decreased female fertility indexDecreased number of corpora lutea in the ovariesDecreased number of implantation sites (or increased pre‐implantation loss)Decreased litter sizeDecreased number of normal oestrus cycles by vaginal smears (during lifetime)Altered oestrus cycle length (during lifetime)Change in the number of ovarian follicles (primordial, primary, secondary and tertiary)Altered maturation stage of ovarian folliclesChange in ovary histopathology Change in ovary weightChange in uterus weightUterus histopathology (with cervix)Change in cervix weightVagina histopathologyVaginal smears at necropsy (OECD TG 443)Oestrus cycle stage at histology done at sacrifice to correlate the stage of cycling with organs (to place in a separate cell) (OECD TG 443 not in 416)Oviduct histopathology 3. Altered Gestation Decreased gestation index (%)Altered duration of gestation (insemination to parturition) 4. Altered Parturition Increased number of females with dystocia (prolonged or difficult parturition) Specific effects and indicators of female reproductive function including fertility. Delayed time to mating Delayed time to mating Reduced female mating index Reduced female mating index Decreased female fertility index Decreased female fertility index Decreased number of corpora lutea in the ovaries Decreased number of corpora lutea in the ovaries Decreased number of implantation sites (or increased pre‐implantation loss) Decreased number of implantation sites (or increased pre‐implantation loss) Decreased litter size Decreased litter size Decreased number of normal oestrus cycles by vaginal smears (during lifetime) Decreased number of normal oestrus cycles by vaginal smears (during lifetime) Altered oestrus cycle length (during lifetime) Altered oestrus cycle length (during lifetime) Change in the number of ovarian follicles (primordial, primary, secondary and tertiary) Change in the number of ovarian follicles (primordial, primary, secondary and tertiary) Altered maturation stage of ovarian follicles Altered maturation stage of ovarian follicles Change in ovary histopathology Change in ovary histopathology Change in ovary weight Change in ovary weight Change in uterus weight Change in uterus weight Uterus histopathology (with cervix) Uterus histopathology (with cervix) Change in cervix weight Change in cervix weight Vagina histopathology Vagina histopathology Vaginal smears at necropsy (OECD TG 443) Vaginal smears at necropsy (OECD TG 443) Oestrus cycle stage at histology done at sacrifice to correlate the stage of cycling with organs (to place in a separate cell) (OECD TG 443 not in 416) Oestrus cycle stage at histology done at sacrifice to correlate the stage of cycling with organs (to place in a separate cell) (OECD TG 443 not in 416) Oviduct histopathology Oviduct histopathology Decreased gestation index (%) Decreased gestation index (%) Altered duration of gestation (insemination to parturition) Altered duration of gestation (insemination to parturition) Increased number of females with dystocia (prolonged or difficult parturition) Increased number of females with dystocia (prolonged or difficult parturition) Changes in clinical chemistry parameters are not considered suitable indicators for identification of the specific effects for female reproductive function due to lack of data since measurements of female reproductive hormone levels are not foreseen in the currently available OECD test guidelines on reproductive and repeated dose short‐term toxicity (see also Section 2.3). However, when such measurements are available, they will be collected only for risk drivers in order to assess whether disruption of the female reproductive system is through an endocrine mode of action or not. Instructions for data collection List of prioritised substances The final list of prioritised a.s. and metabolites will be defined by EFSA according to the methodology described in Section 3.2 and communicated to AGES prior to the data collection. Data collection process Relevant data will be collected by AGES from the toxicological studies with the prioritised a.s. and any metabolites of these substances where the predefined indicators were identified that address reproductive toxicity listed in the sections below. These data refer to any reported indicator of the specific effects defined in Section 4.1 as well as any further information on risk drivers needed to implement efficiently the principles of hazard characterisation (see Section 4.4) and of the elicitation of CAG membership probabilities (see Section 4.5). An ad hoc data collection template has been developed. Data sources The main sources of data are the DARs, RARs generated by rapporteur Member States (RMSs) in the context of Regulation (EC) No 1107/2009 and the respective EFSA conclusions. Additional sources of data (e.g. Joint Meeting on Pesticide Residues (JMPR) evaluations and biocides evaluation by ECHA) will always be checked as additional sources of data; potential differences will be flagged, and additional data (e.g. original study reports) will be collected on a case‐by‐case basis. Data collection for hazard identification and characterisation for establishing CAGs The type of toxicological studies that are considered as appropriate for the purpose of hazard identification and characterisation are all in vivo guideline studies rated in the DARs/RARs or other regulatory assessments as ‘acceptable’ or ‘supportive’, as far as they reliably assess reproductive toxicity parameters in rodents. These include: Reproductive toxicity studies OECD TG 443, OECD TG 416, OECD TG 415, OECD TG 421, OECD TG 422.Repeated‐dose toxicity studies OECD TG 407 (range‐finding study; data will be collected only in case the study is sufficiently powered, i.e. testing at least 10 animals/sex/group), OECD TG 408.Other reproductive or repeated‐dose toxicity studies of acceptable quality according to the DAR/RAR or other regulatory assessments, e.g. US EPA, JMPR, ECHA. Reproductive toxicity studies OECD TG 443, OECD TG 416, OECD TG 415, OECD TG 421, OECD TG 422. Reproductive toxicity studies OECD TG 443, OECD TG 416, OECD TG 415, OECD TG 421, OECD TG 422. Repeated‐dose toxicity studies OECD TG 407 (range‐finding study; data will be collected only in case the study is sufficiently powered, i.e. testing at least 10 animals/sex/group), OECD TG 408. Repeated‐dose toxicity studies OECD TG 407 (range‐finding study; data will be collected only in case the study is sufficiently powered, i.e. testing at least 10 animals/sex/group), OECD TG 408. Other reproductive or repeated‐dose toxicity studies of acceptable quality according to the DAR/RAR or other regulatory assessments, e.g. US EPA, JMPR, ECHA. Other reproductive or repeated‐dose toxicity studies of acceptable quality according to the DAR/RAR or other regulatory assessments, e.g. US EPA, JMPR, ECHA. Additionally, the following will be considered for substances for which no other data are available: Reproductive toxicity studies in guinea pigsReproductive toxicity studies in non‐human primates (NHP), based on expert judgement. Reproductive toxicity studies in guinea pigs Reproductive toxicity studies in guinea pigs Reproductive toxicity studies in non‐human primates (NHP), based on expert judgement. Reproductive toxicity studies in non‐human primates (NHP), based on expert judgement. Data collection for CAG membership probability (for risk drivers only), if applicable A broader data collection will be specifically performed for a.s. or metabolites identified as risk drivers to support the evaluation of the CAG membership probability for each of these substances and for the uncertainty analysis (UA). This supplemental data collection will only take place at a later stage, after setting up the CAGs, conducting the cumulative exposure assessment and identifying the risk drivers. The list of risk drivers and the underlying hazard and exposure‐based methodology will be fully described in a subsequent exposure report (also see Output 3 in Appendix A, Table A.1, step 13). For risk drivers, the conclusions of the EFSA and ECHA's ED assessments will be considered. Besides the studies listed above, in vivo/in vitro mechanistic studies, including mechanistic studies up to 13 weeks of duration in intact animals (e.g. OECD TG 440, OECD TG 441, OECD TG 455, OECD TG 456, OECD TG 458 or other non‐guideline studies) and ToxCast data will be collected. The United States Environmental Protection Agency (U.S. EPA) CompTox Chemicals Dashboard will be consulted for updated screening information on MoA (see Section 3.1) not already included in the regulatory documents above. For risk drivers, the conclusions of the EFSA and ECHA's ED assessments will be considered. Besides the studies listed above, in vivo/in vitro mechanistic studies, including mechanistic studies up to 13 weeks of duration in intact animals (e.g. OECD TG 440, OECD TG 441, OECD TG 455, OECD TG 456, OECD TG 458 or other non‐guideline studies) and ToxCast data will be collected. The United States Environmental Protection Agency (U.S. EPA) CompTox Chemicals Dashboard will be consulted for updated screening information on MoA (see Section 3.1) not already included in the regulatory documents above. For risk drivers, the conclusions of the EFSA and ECHA's ED assessments will be considered. Besides the studies listed above, in vivo/in vitro mechanistic studies, including mechanistic studies up to 13 weeks of duration in intact animals (e.g. OECD TG 440, OECD TG 441, OECD TG 455, OECD TG 456, OECD TG 458 or other non‐guideline studies) and ToxCast data will be collected. The United States Environmental Protection Agency (U.S. EPA) CompTox Chemicals Dashboard will be consulted for updated screening information on MoA (see Section 3.1) not already included in the regulatory documents above. Studies excluded from the data collection Acute oral, dermal and inhalation toxicity studies designed to set LD50/LC50 values.Dermal and inhalation repeated‐dose toxicity studies.Genotoxicity studies (in vitro and in vivo).ADME data (only considered at a later stage for the assessment of predictivity of animal models to human fertility for risk drivers if needed – see Section 2.4.3)Long‐term toxicity and carcinogenicity studies, since reproductive function is not assessed, and gonad histopathology is performed in ageing animals. Endpoints from interim sacrifices (e.g. 1‐year rat) are already covered by short‐term and reproductive toxicity studies.All types of dog studies, due to the following reasons (see also Section 2.3.4): ○Small sample size (n = 4/dose group).○Immature age of dogs at start of 90‐day treatment and pubertal age at the time of sacrifice: sexually immaturity and interindividual variability in the maturation stage may confound the interpretation of histological changes also in controls.○For the 1‐year dog studies similar considerations apply and confounding factors due to the long‐time treatment during the peripubertal period would still be present. Additionally, the 1‐year dog study is not a data requirement under Regulation (EC) 1107/2009. Acute, short‐term and delayed neurotoxicity studies since reproductive function is not assessed as part of the standard OECD test protocol. Acute oral, dermal and inhalation toxicity studies designed to set LD50/LC50 values. Acute oral, dermal and inhalation toxicity studies designed to set LD50/LC50 values. Dermal and inhalation repeated‐dose toxicity studies. Dermal and inhalation repeated‐dose toxicity studies. Genotoxicity studies (in vitro and in vivo). Genotoxicity studies (in vitro and in vivo). ADME data (only considered at a later stage for the assessment of predictivity of animal models to human fertility for risk drivers if needed – see Section 2.4.3) ADME data (only considered at a later stage for the assessment of predictivity of animal models to human fertility for risk drivers if needed – see Section 2.4.3) Long‐term toxicity and carcinogenicity studies, since reproductive function is not assessed, and gonad histopathology is performed in ageing animals. Endpoints from interim sacrifices (e.g. 1‐year rat) are already covered by short‐term and reproductive toxicity studies. Long‐term toxicity and carcinogenicity studies, since reproductive function is not assessed, and gonad histopathology is performed in ageing animals. Endpoints from interim sacrifices (e.g. 1‐year rat) are already covered by short‐term and reproductive toxicity studies. All types of dog studies, due to the following reasons (see also Section 2.3.4): ○Small sample size (n = 4/dose group).○Immature age of dogs at start of 90‐day treatment and pubertal age at the time of sacrifice: sexually immaturity and interindividual variability in the maturation stage may confound the interpretation of histological changes also in controls.○For the 1‐year dog studies similar considerations apply and confounding factors due to the long‐time treatment during the peripubertal period would still be present. Additionally, the 1‐year dog study is not a data requirement under Regulation (EC) 1107/2009. All types of dog studies, due to the following reasons (see also Section 2.3.4): ○Small sample size (n = 4/dose group).○Immature age of dogs at start of 90‐day treatment and pubertal age at the time of sacrifice: sexually immaturity and interindividual variability in the maturation stage may confound the interpretation of histological changes also in controls.○For the 1‐year dog studies similar considerations apply and confounding factors due to the long‐time treatment during the peripubertal period would still be present. Additionally, the 1‐year dog study is not a data requirement under Regulation (EC) 1107/2009. ○Small sample size (n = 4/dose group). Small sample size (n = 4/dose group). ○Immature age of dogs at start of 90‐day treatment and pubertal age at the time of sacrifice: sexually immaturity and interindividual variability in the maturation stage may confound the interpretation of histological changes also in controls. Immature age of dogs at start of 90‐day treatment and pubertal age at the time of sacrifice: sexually immaturity and interindividual variability in the maturation stage may confound the interpretation of histological changes also in controls. ○For the 1‐year dog studies similar considerations apply and confounding factors due to the long‐time treatment during the peripubertal period would still be present. Additionally, the 1‐year dog study is not a data requirement under Regulation (EC) 1107/2009. For the 1‐year dog studies similar considerations apply and confounding factors due to the long‐time treatment during the peripubertal period would still be present. Additionally, the 1‐year dog study is not a data requirement under Regulation (EC) 1107/2009. Acute, short‐term and delayed neurotoxicity studies since reproductive function is not assessed as part of the standard OECD test protocol. Acute, short‐term and delayed neurotoxicity studies since reproductive function is not assessed as part of the standard OECD test protocol. Sub‐acute (oral) studies (e.g. 28‐day studies) unless sufficiently powered (i.e. testing at least 10 animals/sex/group). Developmental toxicity, including developmental neurotoxicity and immunotoxicity, studies where treatment of female rodents occurs only during gestation.All other non‐rodent studies (with the exception of studies in guinea pigs and non‐human primates to be considered only for substances for which no other data are available, see Section 4.2.4). Developmental toxicity, including developmental neurotoxicity and immunotoxicity, studies where treatment of female rodents occurs only during gestation. Developmental toxicity, including developmental neurotoxicity and immunotoxicity, studies where treatment of female rodents occurs only during gestation. All other non‐rodent studies (with the exception of studies in guinea pigs and non‐human primates to be considered only for substances for which no other data are available, see Section 4.2.4). All other non‐rodent studies (with the exception of studies in guinea pigs and non‐human primates to be considered only for substances for which no other data are available, see Section 4.2.4). Criteria for the inclusion of a substance into a CAG The inclusion of an a.s. in a CAG is determined by the presence of one or more indicator(s) of the respective specific effect. Such observations should be treatment‐related, statistically significant and/or biologically relevant (i.e. increase in incidence and/or severity of the observation), present in at least one in vivo toxicological study conducted by the oral route and assessed as « acceptable » and/or « supportive » 5 in the DAR, RAR or equivalent document, unless the observation: occurred only at doses causing excessive systemic toxicity (see ECHA, 2024)is proven to be not relevant for human risk assessment (see Section 2.4 for considerations on human relevance)In the context of this work, it is acknowledged that the definition of the MTD (maximum tolerated dose), commonly taken as a reference point to determine whether an effect is direct or secondary to other systemic effects, is challenging and a matter of debate, since it can vary depending on the endpoint considered and the regulatory framework. occurred only at doses causing excessive systemic toxicity (see ECHA, 2024) occurred only at doses causing excessive systemic toxicity (see ECHA, 2024) is proven to be not relevant for human risk assessment (see Section 2.4 for considerations on human relevance)In the context of this work, it is acknowledged that the definition of the MTD (maximum tolerated dose), commonly taken as a reference point to determine whether an effect is direct or secondary to other systemic effects, is challenging and a matter of debate, since it can vary depending on the endpoint considered and the regulatory framework. is proven to be not relevant for human risk assessment (see Section 2.4 for considerations on human relevance) In the context of this work, it is acknowledged that the definition of the MTD (maximum tolerated dose), commonly taken as a reference point to determine whether an effect is direct or secondary to other systemic effects, is challenging and a matter of debate, since it can vary depending on the endpoint considered and the regulatory framework. As a pragmatic approach, the WG decided to consider the criteria used by ECHA for classifying substances for reproductive toxicity (see Guidance on the Application of the CLP Criteria Version 6.0 – January 2024 https://echa.europa.eu/documents/10162/2324906/clp_en.pdf/58b5dc6d‐ac2a‐4910‐9702‐e9e1f5051cc5, p. 402). Specifically, this Guidance states: ‘All findings on reproductive toxicity should be considered for classification purposes irrespective of the level of parental toxicity. A comparison between the severity of the effects on fertility/development and the severity of other toxicological findings must be performed.’…………….. ‘Adverse effects on fertility and reproductive performance seen only at dose levels causing marked systemic toxicity (e.g. lethality, dramatic reduction in absolute body weight, coma) are not relevant for classification purposes. There is no established relationship between fertility effects and less marked systemic toxicity. Therefore, it should be assumed that effects on fertility seen at dose levels causing less marked systemic toxicity are not a secondary consequence of this toxicity. However, mating behaviour can be influenced by parental effects not directly related to reproduction (e.g. sedation, paralysis), and such effects on mating behaviour may not warrant classification……..’ Consequently, in this context, the WG concurred that an indicator will always trigger the inclusion of a substance into a CAG unless it is observed at doses causing excessive toxicity (e.g. death, severe prostration of the animal, dramatic reduction in absolute body weight, coma, sedation, paralysis). In addition, limitations reported by the authors of the source documents and identified during the data collection and flagged as impacting the validity of the study, should also be considered in order to conclude if such limitations might have an impact on the allocation of a substance into a CAG. Based upon all considerations, a substance can be included in one or several CAGs. Instructions for the hazard characterisation of the substances included in the CAGs This section illustrates the methodology to be followed for characterising each substance included in a CAG. Each active substance or metabolite included in a CAG will be characterised by the assignment of an overall no observed adverse effect level (NOAEL) and an overall lowest observed adverse effect level (LOAEL) for the respective specific effect based on the totality of the available oral toxicity studies. All the indicators are considered equally relevant for the setting of the NOAEL/LOAEL of the respective specific effect, although they may be weighed differently for CAG membership probability (see CAG membership probability under 7). All available information across studies, species, generations and sexes will be considered. The use of sub‐acute studies (e.g. 28‐day studies, preliminary range‐finding reproductive toxicity studies) can be considered to define a NOAEL and a LOAEL only in the case that these studies are sufficiently powered (i.e. with at least 10 animals/sex/group). In each single study, a NOAEL/LOAEL for the specific effect will be identified. In case a study failed to identify a NOAEL and only provided a LOAEL for an indicator(s) of interest, a NOAEL will be derived from this LOAEL by applying a default uncertainty factor (UF) of 10. Although the EFSA guidance on default values to be used in the absence of measured data (EFSA Scientific Committee, 2012) recommends defining the size of such UF on a case‐by‐case basis, the concerned studies will not be reassessed, and, as a systematic approach, the same default UF will be used unless a more specific extra UF is already used in the assessment report. This conservative UF will then be taken into account in the uncertainty analysis. In view of the setting of the overall NOAEL/overall LOAEL for the specific effect, the most sensitive species is identified as the species in which the lowest LOAEL for any indicator has been observed for that specific effect. This lowest observed LOAEL is adopted as the overall LOAEL of the substance. The overall NOAEL of the substance, to be used in the exposure calculations, is the corresponding NOAEL of the same study. Merging of studies to account for dose spacing for the same indicator should only be considered, in the rare case of repetition of studies in the same laboratory which only differ by dose selection (e.g. equivalent quality, comparable time of conduct, experimental design (same OECD TG version), same species, strain, administration oral route (e.g. studies performed by gavage cannot be combined with studies where the substance has been administered in the diet), same indicators), and given that it is properly justified. BMD analysis Once the risk drivers are identified from the cumulative exposure calculations, benchmark dose (BMD) analysis might be applied, if data allow, in the case a refinement of these calculations is needed. This would be the case, for instance, if the total margin of exposure (MOET) 6 at the 99.9th percentile of the exposure distribution is in the range of the threshold for regulatory consideration. Instructions for CAG membership probability Introduction The amount, reliability, relevance and consistency of evidence for causing the specific effects listed in Tables 9 and 10 (Section 4.1) vary between substances, with some being more likely to actually belong to a CAG than others. This can be quantified by assessing the probability that any substance actually causes the specific effect. In this report, this probability is referred to as CAG membership probability. While in principle the CAG membership probability should be assessed for all substances included in a CAG, this process is resource‐consuming (see below) and experience has shown that it has limited impact, since the outcome of a CRA is in most cases driven by a limited number of substances. CAG membership probability will therefore be assessed only for the substances emerging as risk drivers from the cumulative exposure assessment. It should be noted that CAG composition will not be changed as a result of this exercise, even if the outcome is a low probability for the substance to belong to a given CAG. In case some risk drivers show a low CAG membership probability, this will likely result in an overestimation of the risk that will be considered in the UA and risk characterisation steps of the CRA. This probability can be estimated by a WoE assessment consisting of a stepwise process with the following sequence of points: Defining in precise terms the assessment question applicable to each substance included in the CAG.Identifying LoE that are important for assessing whether the substance causes the effect. LoE typically include but are not necessarily restricted to the indicators of the specific effect under consideration. Depending on the specific effect, additional factors contributing to the evidence can be defined.Rating qualitatively the weight of each LoE according to a 5‐degree scale (low, low to medium, medium, medium to high, high). The LoE are assessed by expert discussion with respect to their reliability and relevance to the assessment question. Each LoE is qualified for the strength of its contribution to the probability of the substance causing the effect.Reviewing the information available and listing the available LoE for individual substances included in the CAG.Assessing for individual substances the probability under consideration using the ‘approximate probability scale’ from EFSA's uncertainty guidance (EFSA Scientific Committee, 2018) in Table 11. Defining in precise terms the assessment question applicable to each substance included in the CAG. Defining in precise terms the assessment question applicable to each substance included in the CAG. Identifying LoE that are important for assessing whether the substance causes the effect. LoE typically include but are not necessarily restricted to the indicators of the specific effect under consideration. Depending on the specific effect, additional factors contributing to the evidence can be defined. Identifying LoE that are important for assessing whether the substance causes the effect. LoE typically include but are not necessarily restricted to the indicators of the specific effect under consideration. Depending on the specific effect, additional factors contributing to the evidence can be defined. Rating qualitatively the weight of each LoE according to a 5‐degree scale (low, low to medium, medium, medium to high, high). The LoE are assessed by expert discussion with respect to their reliability and relevance to the assessment question. Each LoE is qualified for the strength of its contribution to the probability of the substance causing the effect. Rating qualitatively the weight of each LoE according to a 5‐degree scale (low, low to medium, medium, medium to high, high). The LoE are assessed by expert discussion with respect to their reliability and relevance to the assessment question. Each LoE is qualified for the strength of its contribution to the probability of the substance causing the effect. Reviewing the information available and listing the available LoE for individual substances included in the CAG. Reviewing the information available and listing the available LoE for individual substances included in the CAG. Assessing for individual substances the probability under consideration using the ‘approximate probability scale’ from EFSA's uncertainty guidance (EFSA Scientific Committee, 2018) in Table 11. Assessing for individual substances the probability under consideration using the ‘approximate probability scale’ from EFSA's uncertainty guidance (EFSA Scientific Committee, 2018) in Table 11. TABLE 11Approximate probability scale for harmonised use in EFSA.Probability termSubjective probability rangeAlmost certain99%–100%Extremely likely95%–99%Very likely90%–95%Likely66%–90%About as likely as not33%–66%Unlikely10%–33%Very unlikely5%–10%Extremely unlikely1%–5%Almost impossible0%–1% Approximate probability scale for harmonised use in EFSA. The present report just addresses points i–iii of the process. Points iv and v concerning the elicitation of the CAG membership probability will be part of a subsequent step of the process covered by a separate report on risk characterisation and uncertainty analysis. Assessment question The assessment question applicable to each substance included in a CAG was agreed as follows: ’What is the probability that a certain substance can lead to any indicator directly related to the given specific effect (CAG)? Lines of evidence and assessment of their weight The WG defined a comprehensive list of LoE applicable to the assessment question and agreed by consensus on the strength to be assigned to each one (see Table 12). This varies based on its estimated contribution to the certainty that a substance is causally linked to the effect under consideration. Although the full list of LoE may not be applicable to all specific effects, for pragmatic reasons, only one list was developed, and this will be applied as appropriate to each specific effect. The assessment of the CAG membership probabilities will be addressed in the report on the uncertainty analysis and the risk characterisation (output 4 in Section 1.1). If necessary, additional or refined LoE will be considered under output 4, accordingly, and the rationale will be provided. TABLE 12Scores agreed for each line of evidence.Line of evidenceStrength (agreed score) 7 ReasoningConcluded as ED for EAS‐modalities by EFSA High If an a.s./metabolite is concluded as an ED for EAS modalities (according to the EFSA/ECHA Guidance), a mechanism of action has been identified increasing the biological plausibility and supporting a causal link between the treatment and an adverse reproductive effect.Harmonised or proposed classification and labelling 8 as Repr. 1A or 1B 9 High If an a.s./metabolite has a harmonised or proposed classification as Category 1 for reproductive toxicity and there is evidence that the classification is driven by the specific effect, this indicates that another independent expert assessment has concluded with a high level of certainty that the substance causes the effect.Notified classification6 as Repr. 1A or 1B High If an a.s./metabolite has a notified classification as Category 1 for reproductive toxicity and there is evidence that the classification is driven by the specific effect, this indicates that another expert assessment (notifier) has concluded with a high level of certainty that the substance causes the effect.Harmonised and proposed classification and labelling6 as Repr. 2 Medium If an a.s./metabolite has a harmonised or proposed classification as Category 2 for reproductive toxicity and there is evidence that the classification is driven by the specific effect, this indicates that another independent expert assessment has concluded with a moderate level of certainty that the substance causes the effect.Notified classification6 as Repr. 2 Medium If an a.s./ metabolite has a notified classification as Category 2 for reproductive toxicity and there is evidence that the classification is driven by the specific effect, this indicates that another expert assessment (notifier) has concluded with a moderate level of certainty that the substance causes the effect.Observation of more than one indicator for the specific effect in the data set High The consistency of experimental observations pointing to the same specific effect in the same study or in different studies with a given a.s./metabolite highly increases the likelihood that there is a causal link between the treatment and the effect.Observation of at least one indicator across different generations (P, F1) in the same multigeneration study (reproducibility across generations) High The consistency of observations across generations supports a causal link between the exposure to the a.s./metabolite and the effect.Observation of the same indicator(s) in independent studies with the same species (reproducibility across studies) High The consistency of observations across studies within the same species highly contributes to the evidence that an a.s./metabolite actually causes the effect.Observation of the same indicator(s) in independent studies with different species (reproducibility across species) High The consistency of observations across studies across different species highly contributes to the evidence that an a.s./metabolite actually causes the effect.Observation of one indicator in association with any ancillary finding supporting the observed indicator Medium‐High Complementary changes contribute to the evidence that an a.s./metabolite actually causes the effect. riasThe strength of this LoE is slightly lower than that provided by 2 indicators being co‐present in the same study (see 5th LoE in this table).Existing mechanistic information, e.g. ToxCast, mechanistic studies Medium or High Factual information on the MoA for the specific effect moderately or highly contributes to the evidence that an a.s./metabolite is actually causing the effect.The substance belongs to a known chemical class of concern (e.g. triazoles) High As triazoles potentially disrupt steroid homeostasis, they have been frequently linked to impairment of male/female fertility and pregnancy.Observation of only one indicator in only one study in the data set Low‐Medium (the nature and severity and incidence of the indicator should be taken into account, as well as the number of studies available in the data set) While any indicator is deemed equally valid for the allocation of an a.s./metabolite into a CAG and for defining the NOAEL/LOAEL for a specific effect, different indicators may bear different weights, for what concerns the probability that an a.s./metabolite has been correctly allocated to a CAG. For some substances, the indicator may be covered by multiple valid reproductive toxicity data sets. In this case, the strength of the evidence coming from a single study should be regarded as ‘low’. On the other hand, if the evidence is supported by an incomplete or a minimum data set, the strength should rather be considered ‘medium’. Evidence of any indicator only at the top dose Medium When demonstrated, a dose–response relationship establishes a causal link between the exposure to the a.s./metabolite and the effect. The lack of investigations at higher doses precludes to proof with high certainty such causal link.Evidence of a clear dose–response relationship for any indicator(s), i.e. increase of the incidence over at least 2 concentrations including the top dose High When demonstrated, a dose–response relationship establishes a causal link between the exposure to the a.s./metabolite and the effect. Scores agreed for each line of evidence. While any indicator is deemed equally valid for the allocation of an a.s./metabolite into a CAG and for defining the NOAEL/LOAEL for a specific effect, different indicators may bear different weights, for what concerns the probability that an a.s./metabolite has been correctly allocated to a CAG. For some substances, the indicator may be covered by multiple valid reproductive toxicity data sets. In this case, the strength of the evidence coming from a single study should be regarded as ‘low’. On the other hand, if the evidence is supported by an incomplete or a minimum data set, the strength should rather be considered ‘medium’. CONCLUSIONS The present report mainly defines the specific effects for male and female reproductive function and the relative indicators (i.e. toxicological endpoints measurable in regulatory studies) for the allocation of pesticides (a.s. and/or metabolites) to the respective CAGs; it is therefore focused on hazard identification. Two specific effects and homonymous CAGs were proposed for male reproduction, namely (1) Altered mating performance and (2) Decreased fertility, while 4 distinct CAGs were suggested for grouping chemicals affecting female reproduction based on (1) altered mating performance, (2) decreased fertility, (3) altered gestation and (4) altered parturition. A list of indicators describing each specific effect was defined as a main criterion that triggers the inclusion of a given a.s./metabolite into the respective CAG. More refined inclusion criteria and a hazard characterisation methodology were elaborated by the WG for being implemented in the next step of the CRA process by AGES. Finally, this report illustrates the LoE for assessing the CAG membership probability, namely the probability that the a.s./metabolites included in the CAGs actually cause the effect of interest. This report was sent for public consultation (see Annex B). Taking into consideration the relevant comments received and further discussions in the WG, the text and the key decisions taken were revisited as needed and finalised for final publication. The actual establishment and characterisation of the CAGs, from the data extraction onwards, will instead be performed in a second step by AGES and published separately. RECOMMENDATIONS General recommendation: In the absence of fully defined and validated AOPs, the proposed specific effects on reproductive function including fertility are exclusively based on adverse outcomes observed in regulatory studies. As AOPs can provide valuable information for the CRA of pesticides altering the reproductive function and in line with the indications of EFSA Scientific Committee (2021), it is recommended to further develop AOPs by focusing on the adverse outcomes that are described in the pesticide data set.Ongoing research on the development of an AOP network (AOPN) on mammalian reproductive toxicity arising from disruption of the EAS modalities is expected to assist the identification of knowledge gaps that can be prioritised for further AOP development (Zilliacus et al., 2024). Initiatives such as the Methods2AOP conceptual data model coordinated by the EU Reference Laboratory for alternatives to animal testing (EURL ECVAM), aim at linking test methods to key events within AOPs and at promoting a broader use of AOP‐based approaches in regulatory risk assessments (Wittwehr et al., 2023). In addition, ongoing research on weight of evidence for cross‐species conservation of androgen receptor‐based biological activity (Vliet et al., 2023) and the advancement of cross‐species extrapolation of the effects of chemicals in regulatory toxicology (Lalone et al., 2021) may facilitate the human relevance assessment of available AOPs. A recent initiative aims at developing a roadmap to define a NAM‐based integrated testing strategy (ITS) to meet DART regulatory requirements (Fritsche, 2024). In the absence of fully defined and validated AOPs, the proposed specific effects on reproductive function including fertility are exclusively based on adverse outcomes observed in regulatory studies. As AOPs can provide valuable information for the CRA of pesticides altering the reproductive function and in line with the indications of EFSA Scientific Committee (2021), it is recommended to further develop AOPs by focusing on the adverse outcomes that are described in the pesticide data set. In the absence of fully defined and validated AOPs, the proposed specific effects on reproductive function including fertility are exclusively based on adverse outcomes observed in regulatory studies. As AOPs can provide valuable information for the CRA of pesticides altering the reproductive function and in line with the indications of EFSA Scientific Committee (2021), it is recommended to further develop AOPs by focusing on the adverse outcomes that are described in the pesticide data set. Ongoing research on the development of an AOP network (AOPN) on mammalian reproductive toxicity arising from disruption of the EAS modalities is expected to assist the identification of knowledge gaps that can be prioritised for further AOP development (Zilliacus et al., 2024). Initiatives such as the Methods2AOP conceptual data model coordinated by the EU Reference Laboratory for alternatives to animal testing (EURL ECVAM), aim at linking test methods to key events within AOPs and at promoting a broader use of AOP‐based approaches in regulatory risk assessments (Wittwehr et al., 2023). In addition, ongoing research on weight of evidence for cross‐species conservation of androgen receptor‐based biological activity (Vliet et al., 2023) and the advancement of cross‐species extrapolation of the effects of chemicals in regulatory toxicology (Lalone et al., 2021) may facilitate the human relevance assessment of available AOPs. A recent initiative aims at developing a roadmap to define a NAM‐based integrated testing strategy (ITS) to meet DART regulatory requirements (Fritsche, 2024). Ongoing research on the development of an AOP network (AOPN) on mammalian reproductive toxicity arising from disruption of the EAS modalities is expected to assist the identification of knowledge gaps that can be prioritised for further AOP development (Zilliacus et al., 2024). Initiatives such as the Methods2AOP conceptual data model coordinated by the EU Reference Laboratory for alternatives to animal testing (EURL ECVAM), aim at linking test methods to key events within AOPs and at promoting a broader use of AOP‐based approaches in regulatory risk assessments (Wittwehr et al., 2023). In addition, ongoing research on weight of evidence for cross‐species conservation of androgen receptor‐based biological activity (Vliet et al., 2023) and the advancement of cross‐species extrapolation of the effects of chemicals in regulatory toxicology (Lalone et al., 2021) may facilitate the human relevance assessment of available AOPs. A recent initiative aims at developing a roadmap to define a NAM‐based integrated testing strategy (ITS) to meet DART regulatory requirements (Fritsche, 2024). Recommendations for consideration by the applicants: Extended one‐generation reproductive toxicity (EOGRT) studies are recommended to be supplemented by a second‐generation (F2) study to investigate reproductive toxicity according to guideline‐compliant studies. These would help clarify the consistency of the effects across generations and in turn strengthen the evidence supporting a causal association between exposure and effect.Complementing the EOGRT (OECD TG 443) and the two‐generation reproductive toxicity (OECD TG 416) studies with hormonal measurements is expected to decrease in the future the uncertainty in the assessment of CAG membership probability and also to facilitate grouping based on MoA data. Extended one‐generation reproductive toxicity (EOGRT) studies are recommended to be supplemented by a second‐generation (F2) study to investigate reproductive toxicity according to guideline‐compliant studies. These would help clarify the consistency of the effects across generations and in turn strengthen the evidence supporting a causal association between exposure and effect. Extended one‐generation reproductive toxicity (EOGRT) studies are recommended to be supplemented by a second‐generation (F2) study to investigate reproductive toxicity according to guideline‐compliant studies. These would help clarify the consistency of the effects across generations and in turn strengthen the evidence supporting a causal association between exposure and effect. Complementing the EOGRT (OECD TG 443) and the two‐generation reproductive toxicity (OECD TG 416) studies with hormonal measurements is expected to decrease in the future the uncertainty in the assessment of CAG membership probability and also to facilitate grouping based on MoA data. Complementing the EOGRT (OECD TG 443) and the two‐generation reproductive toxicity (OECD TG 416) studies with hormonal measurements is expected to decrease in the future the uncertainty in the assessment of CAG membership probability and also to facilitate grouping based on MoA data. Recommendations for consideration by study directors: In relation to toxicological testing improvements, an aspect worthy of special interest is to clearly differentiate reproductive function impairments resulting from a direct chemical insult from those mediated indirectly by parental toxicity.Concerning data reporting, it is highly recommended to use harmonised nomenclature to ensure unambiguous interpretations of the findings and to promote the electronic submission of raw data, in tabulated format, to facilitate data analysis and reuse.The use of BMD modelling to establish the dose eliciting critical effects, rather than the traditional LOAELs and NOAELs, is expected to improve hazard characterisation taking into consideration the shape of the dose response curves and reduce the impact of dose spacing on the definition of the critical doses. In relation to toxicological testing improvements, an aspect worthy of special interest is to clearly differentiate reproductive function impairments resulting from a direct chemical insult from those mediated indirectly by parental toxicity. In relation to toxicological testing improvements, an aspect worthy of special interest is to clearly differentiate reproductive function impairments resulting from a direct chemical insult from those mediated indirectly by parental toxicity. Concerning data reporting, it is highly recommended to use harmonised nomenclature to ensure unambiguous interpretations of the findings and to promote the electronic submission of raw data, in tabulated format, to facilitate data analysis and reuse. Concerning data reporting, it is highly recommended to use harmonised nomenclature to ensure unambiguous interpretations of the findings and to promote the electronic submission of raw data, in tabulated format, to facilitate data analysis and reuse. The use of BMD modelling to establish the dose eliciting critical effects, rather than the traditional LOAELs and NOAELs, is expected to improve hazard characterisation taking into consideration the shape of the dose response curves and reduce the impact of dose spacing on the definition of the critical doses. The use of BMD modelling to establish the dose eliciting critical effects, rather than the traditional LOAELs and NOAELs, is expected to improve hazard characterisation taking into consideration the shape of the dose response curves and reduce the impact of dose spacing on the definition of the critical doses. ABBREVIATIONS Active substance(s) Acceptable Daily Intake Absorption, Distribution, Metabolism and Excretion Anogenital distance Austrian Agency for Health and Food Safety GmbH Adverse outcome Adverse outcome pathway Collaborative Adverse Outcome Pathway Wiki Androgen Receptor Acute Reference Dose Androgen Receptor Trans Activation Benchmark dose Cumulative assessment groups Classification, Labelling and Packaging Cumulative risk assessment Draft Renewal Assessment Report Dihydrotestosterone Danish Technical University Oestrogen, androgen, thyroid, and steroidogenesis European Chemical Agency Endocrine disruptor Extended Advisory Group for Molecular Screening and Toxicogenomics Expert Knowledge Elicitation Extended one‐generation reproductive toxicity Oestrogen Receptor Oestrogen Receptor Trans Activation European Union Framework Partnership Agreement Follicle‐stimulating hormone Guidance Document Gonadotropin‐releasing hormone Glucocorticoid Receptor Hypothalamic–pituitary–gonadal Hypothalamic–pituitary–ovarian Hazard quotient Joint Meeting on Pesticide Residues Key Event(s) Key event relationship Levator ani/bulbocavernosus muscle complex Luteinising hormone Luteinising hormone‐releasing hormone Lowest Observed Adverse Effect Level Line of Evidence Monte Carlo Risk Assessment Molecular initiating event Mode of Action Margin of exposure total Maximum residue levels Maximum Tolerated Dose New approach methodology No Observed Adverse Effect Level Organisation for Economic Co‐operation and Development Ovarian surface epithelium Plant protection product EFSA Panel on Plant Protection Products and their Residues Renewal Assessment Report Rapporteur Member State Reference Point Sertoli cell Test Guideline Terms of reference Toxicity Forecasting Uncertainty Analysis Weight of Evidence REQUESTOR EFSA QUESTION NUMBER EFSA‐Q‐2023‐00734 COPYRIGHT FOR NON‐EFSA CONTENT EFSA may include images or other content for which it does not hold copyright. In such cases, EFSA indicates the copyright holder and users should seek permission to reproduce the content from the original source. Supporting information Annex A: AOP list Annex B: Outcome of the public consultation on the specific effects on the reproductive function including fertility relevant for cumulative assessment of pesticide residues