Following a request from the European Commission (EC), the Scientific Committee (SC) of the European Safety Authority (EFSA) was asked to deliver a scientific opinion on the hazard assessment of endocrine disruptors (EDs). More specifically the SC was asked to advise on i) the scientific criteria to distinguish between EDs and other groups of substances with different modes of action, ii) the criteria to distinguish between physiological modulation and adverse effects on humans and on the ecosystem as a result of exposure to endocrine active substances (EASs), and iii) to review existing test methods and discuss their appropriateness for the identification and characterisation of effects mediated by EASs.
The SC defines an EAS as a substance having the inherent ability to interact or interfere with one or more components of the endocrine system resulting in a biological effect, but need not necessarily cause adverse effects. Therefore, the SC considers endocrine activity as a collection of modes of action, potentially leading to adverse outcomes, rather than a (eco)toxicological hazard in itself. The SC endorsed the WHO/IPCS 2002 definition of an ED implying that there must be reasonable evidence for a biologically plausible causal relationship between the endocrine activity and the induced adverse effect(s) seen in an intact organism or a (sub)population for a substance to be identified as an ED.
Assessment of adversity is not unique to endocrine related effects. Scientific criteria for assessment of adversity have not been generally defined. In general, but not always, transient, inconsistent and minor fluctuations at the biochemical and molecular level may be considered adaptive, i.e. non-adverse. Changes at the cell-, organ-, organism-, or (sub)population-level resulting in pathology or functional impairment in vivo, as well as altered timing of development, may be considered adverse. It is therefore difficult to propose ED-specific criteria for adversity and expert judgement in a weight-of-evidence approach is needed to assess substances for possible endocrine disrupting properties.
The Organisation for Economic Co-operation and Development (OECD) revised Conceptual Framework (CF) provides a guide to the data sources, the OECD test guidelines and standardised test methods available, under development or proposed for the evaluation of chemicals for EASs/EDs. Information on endocrine activity can be obtained from existing information, read-across, in silico tools, in vitro and in vivo screening assays (Levels 1, 2 and 3 of the CF), or from other mechanistic investigations. A prerequisite for an EAS to be regarded as an ED is the need to identify the adverse effect. For this, test methods with apical endpoints (Levels 4 and 5 of the CF) can be used together with existing information, read-across and other in vivo (eco)toxicity tests that provide information on apical endpoints. Therefore, in principle, no single assay is likely to provide all the information needed to decide whether a substance is an ED because of the need to provide both mechanistic and apical information.
Taken together, but bearing in mind the recommendations made in this opinion, a reasonably complete suite of standardised assays (for testing the effects of EASs) is (or will soon be) available for the oestrogen, androgen, thyroid, or steroidogenesis (EATS) modalities in mammals and fish, with fewer tests available for birds and amphibians. While downstream effects of disruption of some non-EATS pathways/modalities may be detectable in some of the standardised apical vertebrate assays, it is important to recognise that standardised mechanistic assays for non-EATS modalities relevant to mammals, fish and other vertebrates are not or not yet available. For invertebrates, standardised mechanistic assays are lacking from the OECD testing suite, mainly due to poor understanding of invertebrate endocrinology. Finally, a range of major taxa, e.g. reptiles or echinoderms have not yet been considered by OECD for any endocrine assay development. It is unknown at present whether it will be possible to read-across to untested groups from tests with other taxa.
The SC identified the need for further development of screens and test methods, particularly with regard to non-EATS modalities that may be associated with adverse effects in humans or the environment.
The SC discussed a number of general aspects related to the testing of substances (independently of whether these are EASs or substances with other modes of action):
- Exposure of organs and tissues to certain substances at critical point(s) during their development can result in irreversible change of the organ/tissue. The SC noted that, although some of the tests in the OECD CF do cover exposure during critical periods of development in utero, current mammalian tests may not cover effects that might be induced by exposure during fetal or pubertal development, but may emerge during later life stages. Fish lifecycle tests cover all relevant windows of exposure and can be expected to reveal the longer-term effects of developmental exposures at all stages of the lifecycle.
- The SC recognises that combined exposure to multiple EASs could occur in such a way that combined toxicity could arise. The issue of combined toxicity resulting from combined exposure to multiple substances will be addressed by EFSA in a separate activity.
- The SC noted the lack of consensus in the scientific community with regard to the existence and/or relevance of low-dose effects and non-monotonic dose response curves (NMDRCs) in (eco)toxicology in connection with endocrine activity, endocrine disruption or other endpoints/modes of actions.
The SC recommends as a follow up activity to clarify in a broader context the issues of biological thresholds and criteria for adversity, combined exposure to multiple substances and NMDRCs.
The SC also underlines the need for the further development of testing strategies to generate adequate data for the identification and assessment of endocrine disrupting properties. An example has been developed in outline for fish species by the OECD.
The SC discussed several aspects that can be considered for hazard characterisation of EDs. The SC is of the opinion that hazard characterisation (e.g. establishment of a health-/ecotoxicology-based guidance value) should be based on the effect leading to the lowest health/ecotoxicology-based guidance value, irrespective of the mode of action. Such a health/ecotoxicology-based guidance value would also protect against endocrine-mediated effects occurring at higher doses. With regard to the use of severity, (ir)reversibility and potency for the hazard characterisation of EDs, the SC considers that to inform on a level of concern for endocrine disrupting substances, these elements should be evaluated in relation to the degree, duration and timing of exposure. Levels of concern are not determined exclusively by risk assessment but also by protection goals set by the risk management.
In conclusion, EDs, of natural or synthetic origin, can be identified according to three criteria: endocrine activity, adversity of effects and a plausible link between endocrine activity and adverse effect. The SC considers that a reasonably complete suite of assays is (or will soon be) available to identify and characterise the important hazards of EATS substances in mammals and fish, with fewer tests available for birds and amphibians. Furthermore, these evaluation methods should, in principle, be fit for the purpose of establishing safe doses/concentrations of EDs if (1) certain aspects (e.g. follow up of exposure in critical windows of susceptibility to later life stages) are addressed and (2) used with all available information in a weight-of-evidence approach. It should also be noted that standardised mechanistic assays for non-EATS modalities relevant to mammals, fish and other vertebrates are not yet available. For invertebrates, relevant mechanistic assays are lacking from the OECD testing suite. Finally, a range of major taxa e.g. reptiles or echinoderms, have not yet been considered by the OECD for any endocrine assay development.
Furthermore, to inform on risk and level of concern for the purpose of risk management decisions it is the opinion of the SC that risk assessment (taking into account hazard and exposure data/predictions) makes best use of available information. EDs can therefore be treated like most other substances of concern for human health and the environment, i.e. be subject to risk assessment and not only to hazard assessment.