Can we predict and manage the risks which endocrine disrupting chemicals pose for wildlife?

Peter Matthiessen,
Independent Consultant Ecotoxicologist,
United Kingdom

Peter Matthiessen

Can we predict the risks to wildlife of endocrine disrupting chemicals (EDCs)? Until very recently, the answer to this question was certainly 'no!'. Almost none of the ecotoxicity tests which were routinely used to evaluate the environmental safety of new and existing substances were sensitive to the unique effects of EDCs, and as a consequence, these chemicals entered the market without any specific regulation and some caused serious damage to certain wildlife populations. One classic example observed in Japan and elsewhere concerned the antifouling paint ingredient tributyltin (TBT) which was widely used on ships' hulls and caused many mollusc populations to crash by interfering with their hormone systems and consequently damaging breeding success1. Other such cases involved estrogens like the alkylphenol ethoxylate surfactants and the synthetic hormones in oral contraceptives, which mimicked the hormone estradiol and caused (and are still causing) the appearance of sexually and reproductively compromised (intersex) fish in many rivers and estuaries around the world2. There are many other examples which have now been documented to a greater or lesser extent. The point is that such effects were not suspected and therefore not looked for during chemical development or registration.

These findings prompted a huge programme of research into the effects themselves, and into improved ecotoxicity testing methods which would be able to identify and characterise them. The test development and standardisation work was coordinated by the Organisation for Economic Cooperation and Development (OECD) which is responsible for publishing ecotoxicity testing methods that have broad international acceptance3. Although much work remains to be done, we now have in place a robust suite of tests with good sensitivity to the effects of four major classes of EDCs (estrogens/anti-estrogens; androgens/anti-androgens; thyroid system disrupters; and steroidogenesis disrupters) in vertebrates ranging from fish to mammals. Work is continuing to develop further tests. One of the on-going development programmes is working on some promising mollusc-based methods which are very sensitive to substances such as TBT, and methods based on other invertebrates such as insects and crustaceans are already, or soon will be, available. Tests that do not require the use of animals would of course be preferable for ethical reasons, and these are being developed, but at present it is only possible to confirm the presence and effects of an EDC through the humane use of intact living organisms.

After running suitable toxicity tests, the next step with a normal chemical is to establish a Predicted-No-Effect-Concentration (PNEC) which is expected to be harmless to all organisms, using appropriate safety factors applied to the toxicity data to account for various uncertainties (such as extrapolation to protect the many species which have not been tested). The PNEC is then compared with the worst-case Predicted Exposure Concentration (PEC) which is likely to be found in the environment, leading to a risk assessment. If the risk is low (i.e. the PEC is lower than the PNEC), a chemical will generally be permitted for use, perhaps after practical controls to minimise environmental concentrations. The big question is whether this type of risk assessment can be applied to EDCs.

There are a number of particular properties possessed by many EDCs which, it has been argued, imply that risk assessments may be unsound. First of all, some EDCs are able to cause profound and irreversible effects on organisms following brief exposure in the uterus or during larval stages. It is chemicals of this type to which the older (eco)toxicity tests were often insensitive, but the new suite of methods published by OECD is able to detect them. For example, the OECD Fish Sexual Development Test (OECD TG 234) is able to identify permanent effects on sex ratio caused by relatively brief exposure of fish fry. Delayed risks of this type are therefore accounted for by the new testing procedures.

Secondly, it has been claimed that there is no such thing as a safe 'threshold' concentration below which an EDC does not cause effects, but the evidence for this is weak and it appears to be more of a theoretical concern than a real one for most, if not all, EDCs. Thirdly, it was thought that toxicity tests operated at high chemical doses or concentrations would not be able to predict effects at lower concentrations, but the new OECD methods permit the thorough testing of doses/concentrations covering a much larger span than formerly, and the existence of so-called non-monotonic dose-responses remains controversial and unproven.

These and other uncertainties have prompted the European Union (EU) to take a precautionary stance and implement so-called hazard-based regulation of EDCs i.e. to abandon risk assessment. The precise regulatory definition of an EDC is still under discussion in Europe, but the consequence of identifying a pesticide, for example, as an EDC mean it will not be approved for use in the EU, or will be subject to replacement if it is already on the market. This regulatory action is likely to lead to the withdrawal of a number of useful plant protection products, and loss of the ability to grow certain crops economically in particular areas.

There is no doubt that EDCs possess some unusual properties which will need to be considered during the regulatory process. Indeed, it may well be justified to make some modifications to the risk assessment paradigm to take these properties into account, such as the implementation of weight-of-evidence evaluations. However, in my view, abandoning risk assessment of EDCs altogether is a serious over-reaction which may have dangerous knock-on consequences far beyond the field of endocrine disruption. The EU certainly appears isolated in this respect, because it is expected that both the United States and Japan will evaluate EDCs using risk assessment methods.

As matters stand with chemicals regulation in the EU, we appear to be taking a retrograde and unscientific step equivalent to, for example, banning the use of electricity because of its inherent electrocution hazard. Instead, I believe the mature approach is to evaluate the risk and use appropriate insulation!



1 P. (2013). Detection, monitoring and control of tributyltin ? an almost complete success story. Environmental Toxicology and Chemistry 32 (3).

2 Sumpter, J.P. and Jobling, S. (2013). The occurrence, causes, and consequences of estrogens in the aquatic environment. Environmental Toxicology and Chemistry 32 (2), 249-251.

3 Draft documents for public comments | Organisation for Economic Co-operation and Development