Opinion on BKH Consulting Engineers Report "Towards the establishment of a priority list of substances for further evaluation of their role in endocrine disruption" - Opinion adopted at the 17th CSTEE plenary meeting, Brussels, 5 September, 2000.

Context

In December 1999, the Commission adopted a Communication to the Council and European Parliament on a Community Strategy for Endocrine Disrupters. The Communication identifies a number of short-, medium- and long-term actions to address the problem of endocrine disruption. One of the short-term actions is the establishment of a priority list of substances for further evaluation of their role in endocrine disruption.

This list is intended to be used, inter alia:

- to identify substances for 'priority" testing once agreed test methods become available,

- to identify substances which can be, or are already being addressed, under existing Community legislation covering hazard identification, risk assessment and risk management,

- to identify gaps in knowledge on aspects such as dose/response relationships, sources/pathways of exposure and epidemiological studies of cause/effect relationships which will help guide further research and/or monitoring efforts, and

- to identify specific cases of consumer use, for example, the case of potentially more vulnerable groups of consumers such as children, for special consideration from a consumer policy point of view.

The first step in this process, a study entitled "Towards the establishment of a priority list of substances for further evaluation of their role in endocrine disruption" was carried out by BKH Consulting Engineers, Delft, The Netherlands in association with TNO Nutrition and Food Research, Zeist, The Netherlands.

Terms of reference

The Committee on the basis of examination of the following report:

· Towards the establishment of a priority list of substances for further evaluation of their role in endocrine disruption - preparation of a candidate list of substances as a basis for priority setting, BKH report M0355008/1786Q final with annexes 1 to15, Delft, 21 June 2000,

CSTEE/2000/12 Add. 1 with Add. 1A to 1O

is to answer the following two questions

1. Are the source material, methodology and selection criteria used to include substances in the consultants' list logical and scientifically relevant?

2. Does the CSTEE consider that the preliminary evaluation of substances presented in the study report is an appropriate scientific basis on which to establish a priority list of substances for further evaluation of their role in endocrine disruption? If not, what other methodology and/or evaluation would the CSTEE suggest, bearing in mind the context described above?

The CSTEE has in addition to the BKH report, considered the submitted comments from the European Chemical Industry Council-Endocrine Modulator Study Group (CEFIC-EMSG, CSTEE/2000/12 - Add. 2). The CSTEE has also made use of the Opinion of the Scientific Committee on Plants (SCP) on Endocrine Disruption relevance in the context of Council Directive 91/414/EEC concerning the placing of plant protection products on the market (Opinion expressed by the SCP on 2 December 1999). Further, reference is given to the previous CSTEE Opinion on Human and Wildlife Health Effects of Endocrine Disrupting Chemicals, with Emphasis on Wildlife and on Ecotoxicology Test Methods, expressed on 4 March 1999. The CSTEE has also received information on the classification and labelling on 66 prioritised substances from the EC Joint Research Centre (CSTEE/2000/12 - Add. 5A). In addition, the European Crop Protection Associations has submitted a position paper on the BKH report (CSTEE/2000/12 - Add. 4/b) and CANTOX Health Sciences International has provided the CSTEE with documentation on the thyroidal effects of resorcinol (CSTEE/2000/12 - Add. 3).

Summary Opinion

a. Are the source material, methodology and selection criteria used to include substances in the consultants' list logical and scientifically relevant?

The CSTEE supports a stepwise approach for the selection of substances for further evaluation and prioritisation. However, the CSTEE does not find the source material, methodology and selection criteria used to be scientifically adequate.

b. Does the CSTEE consider that the preliminary evaluation of substances presented in the study report is an appropriate scientific basis on which to establish a priority list of substances for further evaluation of their role in endocrine disruption? If not, what other methodology and/or evaluation would the CSTEE suggest, bearing in mind the context described above?

The Committee concludes that there are important shortcomings in the present approach. Especially, this relates to the omission of dose-response/potency considerations and the inclusion of too restrictive persistence and production volume criteria in the second step of the selection process. Further, more quantitative exposure information should have been included in the second step. Another important omission is that synthetic hormones released into the environment have not been considered. Also, substances lacking data have not been properly addressed. Several suggestions for improvement of the current approach are presented in the following comments to the study report.

CSTEE Comments to BKH Report

Introduction

In its Opinion of 4 March 1999, the CSTEE recognised the growing concern on possible harmful consequences of exposure to xenobiotic compounds that are capable of modulating the endocrine system and thus have the potential to adversely affect human and wildlife reproductive health. At present, this concern has not been substantiated with respect to human health in that no causative role for endocrine disrupting chemicals in diseases and abnormalities possibly related to an endocrine disturbance has been verified. However, the CSTEE re-commended to further evaluate the human health effects that have been associated with endocrine disrupters, and to identify the underlying causes. On the other hand, impaired reproduction and development of several wildlife species have been causally linked to exposure to endocrine disrupting chemicals and have resulted in local or regional population changes. Thus, the European Commission Community Strategy for Endocrine Disrupters is seen as a timely initiative to address the public concern.

A number of regulatory agencies, institutions and public interest groups have produced lists on endocrine disrupting chemicals (CSTEE/2000/12 Add. 1B). It is important to recognise that the compilation of such lists has been performed with varying levels of scientific input and competence, so that the use of these lists for priority setting and further risk assessment activities must be done with considerable caution. Also, the appropriateness of the term "list" should be carefully considered, it gives a connotation of a quality controlled regulatory instrument. The CSTEE, therefore, questions the advisability of using the term "list" in the present context, even if this list is only preliminary and is meant to undergo considerable further discussion and change before a final list is developed. The CSTEE would have preferred that the term "compilation of selected substances" had been used, rather than the term "priority list".

It is important to realise that endocrine disruption is not a toxicological endpoint per se as is cancer or allergy, but that it is a descriptor for a functional change that may lead to adverse health effects (IPCS definition). Thus, CSTEE strongly warns against the development of endocrine disruption as such as a classification category, used for example in labelling. Rather, endocrine disruption should be seen in the context of well-established endpoints, primarily reproductive toxicity and impaired development. Notwithstanding the definition of an endocrine disrupter, the CSTEE cautions against confusing a secondary effect on endocrine disruption caused by a primary damage of hormone producing organ with a primary alteration of the function of an endocrine system leading to adverse health effects.

When discussing the action of endocrine disrupters, it is imperative to take dose-response and potency aspects into consideration. Only in situations where the endocrine disrupting effect is critical (i.e. is the most potent) in comparison to other toxic effects, should the endocrine disruption be considered for hazard and risk assessments. Little importance should be assigned to situations where the no-observed-effect level for endocrine disruption is substantially higher than that for other adverse effects caused by a chemical.

General views on BKH report

From the review of existing lists and other sources of information, BHK has identified 564 substances with various levels of suspicion of endocrine disruption (Step 1). This number was reduced to 147 (Step 2) after including selection criteria of high production volume (HPV) or high persistence (throughout the report these criteria are erroneously described as "HPV and/or highly persistent"). In Step 3 an expert meeting was used to rank these 147 substances into three categories of decreasing evidence of endocrine disruption, with 66 chemicals in Category 1, 52 chemicals in Category 2 and 29 chemicals in Category 3. In the last step (Step 4), the 66 Category 1 chemicals were reduced to 60 substances (29 clustered substances) after incorporating data relevant to exposure concerns.

The primary criticism of this approach relates to the total omission of potency considerations throughout, and the restricted criteria used in the second step. An initial evaluation of dose-response considerations relative to potential exposures should have been carried out in order to focus more on substances were there is a basis for concern, rather than simply using the mostly qualitative information as selection criteria from the first to the second step.

It is probable that chemicals identified in the first step have been excluded from further prioritisation because they have not fulfilled the HPV and high persistence criteria in the second step. Incorporation of data on use pattern, environmental levels and other quantitative exposure information should have been undertaken at this step.

Both in the second and third step have substances lacking data been given low priority. In Step 2 there were as many as 212 substances lacking data on persistence, these were assigned low priority and not carried forward in the process. In Step 3 there were 18 substances with no or insufficient data on endocrine disrupter-related effects. No mechanism is presented on how these data gaps should be remedied and how these substances should be brought back in the prioritisation process. Also, no mechanism for incorporating new information is presented. The CSTEE recommends that there should be an iterative process both related to data gaps and new information, rather than a one-off linear process.

Objectives and scope of BKH project

The objectives of the current project are stated to be:

- to identify selection criteria and produce a working list of substances associated with endocrine disruption

- to quantify the production volumes and to identify the sources/uses and pathways of human and wildlife exposure for these chemicals

- to prepare a candidate list grouped according to available information on selection criteria

In general, CSTEE supports these objectives and finds that a compilation of candidate substances for further assessment and prioritisation is warranted. However, much more emphasis should have been given to potency considerations and quantification of exposure early in the process.

The project has distinguished two classes of endocrine disrupters, namely: 1) 'Natural' hormones and 2) man-made substances. The latter class has been subdivided into: a) synthetically-produced hormones and b) man-made chemicals. BKH has focussed on the class 2b (man-made chemicals). The CSTEE considers that the subdivision of the second class is artificial (both classes are "man-made"), and that it would have been necessary to include synthetically-produced hormones present in the environment at elevated, non-physiological levels in the current project.

Project approach - steps 1-4

In the first step the BKH has set up an inventory of available lists of potential endocrine disrupters and suspected endocrine disrupters from various organisations. In addition, BKH has collected literature from key experts, review documents and performed a literature search to include the most recent references not covered in the review documents. All of the substance information was collected in a database.

It is obvious that this first step primarily is based on suspicion of endocrine disrupter effects, not on a detailed scrutiny and evaluation of the individual substances entered in this initial compilation. However, this is seen as acceptable as long as the limitations in the scientific quality of this step is recognised and understood. In order to initiate a prioritisation process, one has to start from somewhere, even if there are clear scientific deficiencies in the input data. It is not clear how the BKH included candidates for consideration from primary literature data.

Before the selection criteria for the second step were decided, a stakeholder meeting with representatives of governments, industry and NGOs was held. This meeting recommended the use of high production volume and persistence as criteria for narrowing down the chemicals selected in the first step. A cut-off of production volume of 1000 tonnes per year was applied. The CSTEE finds this criterion too simplistic leading to the possible exclusion of important substances for further prioritisation.

In its Opinion of 4 March 1999, the CSTEE recognised the growing concern on possible harmful consequences of exposure to xenobiotic compounds that are capable of modulating the endocrine system and thus have the potential to adversely affect human and wildlife reproductive health. At present, this concern has not been substantiated with respect to human health in that no causative role for endocrine disrupting chemicals in diseases and abnormalities possibly related to an endocrine disturbance has been verified. However, the CSTEE re-commended to further evaluate the human health effects that have been associated with endocrine disrupters, and to identify the underlying causes. On the other hand, impaired reproduction and development of several wildlife species have been causally linked to exposure to endocrine disrupting chemicals and have resulted in local or regional population changes. Thus, the European Commission Community Strategy for Endocrine Disrupters is seen as a timely initiative to address the public concern.

The CSTEE also finds the application of the persistence criterion to be too restrictive. First of all, only models have been used for assessing persistence. Only substances that take more than months to degrade and with a biodegradation probability of <0.1 were considered to be highly persistent and thus carried forward in the process. This criterion would exclude several substances categorised as persistent organic pollutants (POPs). Using the example of contamination of sewage effluent waters with natural oestrogenic hormones, it is obvious that non-persistent, biodegradable substances can constitute endocrine disruption effects in the environment. For many of the substances under review, there exist test data on bioaccumulation potential and biodegradation, this should have been considered in the second step. Further, it would have been advisable to include field data on persistence.

In the third step, a panel of experts on endocrine disruption was asked to evaluate all available information in order to categorise the selected group of substances. The experts in human health and wildlife used the established database in their evaluation. They evaluated all the data, but more weight was given to positive data. The following criteria were used to categorise the selected chemicals:

Category 1: At least one study providing evidence of endocrine disruption in an intact organism. Not a formal weight of evidence approach

Category 2: Potential for endocrine disruption. In vitro data indicating potential for endocrine disruption in intact organisms. Also includes effect in vivo that may, or may not, be ED-mediated. May include structural analyses and metabolic considerations

Category 3: No scientific basis for inclusion in list or no data

The two types of substances in Category 3 were combined. The CSTEE considers this to be unfortunate, since substances with no or insufficient data, although limited in number (18), are not brought back into the review process. However, the CSTEE finds the general concepts and conduct of the third step to be reasonable and acceptable. It is obvious that a relatively short meeting of this kind could not perform any detailed scrutiny of the primary literature. Rather the meeting was primarily making use of the experts' competence and familiarity with the subject area. Thus, this meeting was more an evaluation by scientists than an in-depth, scientific evaluation.

In the fourth step, Category 1 chemicals from the third step were further evaluated taking exposure information into consideration. However, this information was mostly qualitative. High concern was given to situations were human exposure is expected, due to environmental concentrations and those in food or consumer products, also considering vulnerable groups such as infants and patients. High concern was also given to instances where wildlife exposure is expected, such as for sediment-living organisms and top predators, due to use and emission patterns, and that the chemical is persistent and bioaccumulative.

This last step was not very effective in the prioritisation process, since it only brought down the number of 66 substances from the third step to 60 after the fourth step. The CSTEE criticises this stage of the evaluation in that not more quantitative exposure information was brought into the process. However, as stated before, the CSTEE feels that the more detailed and quantitative exposure information should have been brought in earlier at the second step. It is unfortunate that existing, but unpublished, exposure and effect data on chemicals such as plant production products, were not made available to the present project.

Comments to selected substances

The selection process has resulted in a compilation of 60 substances belonging to 29 clusters, for further evaluation. It is obvious that many of these substances are regulated, in fact the overview presented by the Joint Research Centre finds that most substances are, or in the progress of being, classified in relation to Council Directive (EEC) 67/548 (apart from some PCB congeners and mixtures, dioxin congeners and several organotin substances). Further, several of the substances in the compilation are pesticides regulated by the Council Directive 91/414/EEC concerning the placing of plant protection products on the market.

The CSTEE finds that the present compilation is an inadequate starting point for further hazard and risk assessment, since it is probable that substances that should have been selected, were missed due to the deficiencies in the second step noted above. In the further evaluation it will become apparent that there exists an extensive database and a thorough evaluation process on plant protection substances such as vinclozolin, allowing for a satisfactory assessment of mammalian reproductive and developmental hazards. Further, it will become apparent that endocrine modulating effects such as alterations in thyroid hormone homeostasis, will not necessarily be critical for some of the compiled pesticides compared to other toxic effects caused by these substances. The compilation of selected substances and the subsequent in-depth evaluation of their role in endocrine disrupter-mediated human and environmental effects will be a continuing process. It will be important to make use of all information resources and research outputs in this process, including those from relevant producer and user organisations.

Opinion of SCP

The Scientific Committee on Plants (SCP) concluded in its Opinion of 2 December 1999 that the committee is following the scientific progress in the knowledge about endocrine disruption with attention. However, it does not consider this problem to be of great concern for the assessment of plant production products, because of the current process of evaluation. This permits a rather comprehensive appreciation of the endocrine disruption-related toxicological risk for mammalians and man. Also, ecotoxicological risks arising from endocrine disruption was felt to be generally captured by the current assessment scheme, although for some species (in particular invertebrates) the test programme is not yet satisfactory. Further, the SCP considers it appropriate to wait for the conclusion of the ongoing endocrine disruption test guideline-update and development programme by OECD, before recommending to the EU to undertake specific actions aimed at introducing supplementary testing of pesticides. The CSTEE supports the views presented by the SCP. The European Crop Protection Association (ECPA) concludes that Directive 91/414/EEC provides a comprehensive approach for the human and ecological safety evaluation of plant protection products. However, the CSTEE notes that there are deficiencies in the application of the current ecotoxicological program regarding testing for potential endocrine disrupting effects of plant protection products.

CEFIC Comments

The European Chemical Industry Council (CEFIC) has through its Endocrine Modulator Steering Group (EMSG) submitted substance-specific exposure and effect data on the following 21 substances/groups of substances compiled in the Stage 4: vinclozolin, maneb, metam natrium, thiram, zineb, lindane, linuron, atrazine, acetochlor, alachlor, styrene, butylbenzylphthalate, hexachlorobenzene, dioctyl-phthalate, di-N-butylphthalate, bisphenol A, polybrominated biphenyls, tributyltin compounds, fentin acetate, 3,4-dichloroaniline and resorcinol. The CSTEE finds these data to be very valuable and should prove to be of considerable usefulness in the further evaluation process. However, CSTEE notes that the EMSG in some instances has been selective in its use of referenced documentation (e.g. for bisphenol A). The CEFIC/EMSG has in its comments presented a weight of evidence approach to prioritising action in relation to endocrine disruption. The principles presented in this approach are based on sound evaluation practices and are in line with CSTEE approaches to chemical hazard and risk assessment.

Conclusions

Considering the general scientific issues involved in the area of endocrine disruption and evaluating the BKH Consulting Engineers Report "Towards the establishment of priority list of substances for further evaluation of their role in endocrine disruption", the CSTEE concludes the following:

General Conclusions

1. The use of the term "priority list" gives a connotation of a quality controlled regulatory instrument. The CSTEE would have preferred using the term "compilation of selected substances".

2. Endocrine disruption is not a toxicological endpoint per se as is cancer or allergy. The CSTEE strongly advises against the development of endocrine disruption as such as a classification category, used for example in labelling.

3. Only in situations where the endocrine disrupting effect is critical in comparison to other toxic effects, should endocrine disruption be considered for hazard and risk assessments. Little importance should be assigned to situations where the no-observed-effect level for endocrine disruption is substantially higher than that for other adverse effects caused by the chemical.

Specific Conclusions

1. The CSTEE finds that the approach used by the BKH in selecting substances for further hazard and risk assessment is inadequate. The primary criticism of this approach relates to the total omission of dose-response/potency considerations.

2. The present approach was a one-off linear process for selection of substances for further evaluation. The CSTEE recommends instead applying an iterative process addressing both data gaps and incorporating new information.

3. The CSTEE considers that synthetically-produced hormones should have been included in the project.

4. Although the CSTEE is in agreement with the general objectives of a stepwise approach, the Committee especially finds the selection criteria used in the second step as being too restrictive. Thus, chemicals which should have been included for further evaluation, presumably have been missed.

5. The use of a single cut-off value for production volume is not recommended, since the release of chemicals into the environment is highly dependent on use patterns. Quantitative exposure data, such as from field studies, should have been included in the selection process in the second step. It is unfortunate that existing, but unpublished, exposure and effect data on chemicals such as plant protection products, were not made available to the present project.

6. The persistence criteria applied will exclude some chemicals which have been categorised as persistent organic pollutants. Further, it is obvious that the continuous release of non-persistent, biodegradable substances into the environment, can lead to endocrine disruption effects.

7. The CSTEE finds that it would have been better not to include two types of chemicals in the Category 3 substances of the third step, namely substances for which there was no scientific basis for inclusion and for those which there were no or insufficient data. The latter category, although consisting of a limited number, should have been brought back into the review process.

8. The fourth step in the selection was not very effective, in that it only reduced the number of selected substances from 66 to 60.

9. There are a number of chemicals among the selected 60 substances for which a comprehensive risk assessment has been done or is in progress, such as for plant protection products (91/414/EEC)), biocides (98/8/EEC) and high-production volume chemicals (793/93/EEC). The CSTEE recommends that the need for further evaluation of their role in endocrine disruption should be decided on the basis of the outcome of these risk assessments, under the presumption that due consideration of appropriate endpoints have been undertaken.

10. In the further evaluation process, exposure and effect information on selected substances from relevant producer and user organisations, should be included.

11. In view of the critical comments to the present report, the CSTEE recommends that the Committee should be contacted for inputs to the briefing of external consultants before they begin their task. This is in line with the recommendation from the Scientific Steering Committee on this issue.