TABLE 4.

Where Do We Go from Here?

1. Ecosystem health

1.a. Better characterization of the extent and population impacts on fish of estrogenic municipal effluents, and further research concerning the possible impacts on fish and wildlife of other known EDCs entering the environment (e.g., from pulp and paper mills, CAFOs).

1.b. Clearer delineation of exact chemicals responsible for effluent estrogenicity (i.e., ethinyl estradiol vs. E2 vs. nonylphenol) as a basis for identifying mitigation options.

1.c. Identification of pharmaceuticals that could possibly cause ecological effects through properties similar to EDCs, that is, high-potency chemicals focused on biological pathways (other than the HPG axis) that have a substantial degree of phylogenetic conservation.

2. EDC screening

2.a. Evaluation of the EPA T1S assay suite to determine “holes” and redundancies with respect to identifying HPG-active chemicals (i.e., we aren't going to be able to apply the full suite of tests to very many chemicals-too expensive); an understanding of endpoint/species extrapolation will be critical here.

2.b. Improve predictiveness of screens and tests for EDC activities.

2.c. Enhance and expand the in vitro assays for T1S to replace animal-derived proteins with recombinant receptors, adapt in vitro assays for high throughput.

2.d. Determine the need for in vitro assays using nonmammalian pathways and develop the assays for screening, as warranted.

2.e. Develop an in utero lactational screening assay for possible inclusion in T1S.

2.f. Implement a T1S battery that meets or exceeds the EDSTAC 1998 recommendations.

3. Multigenerational and transgenerational testing including a postnatal examination of offspring

3.a. Enhanced multigenerational assays that include sensitive and reliable end points, that use animals more efficiently (fewer litters/dose, examine more pups/litter, increase histopathological examinations to include all F1 animals) without losing statistical power.

4. Clearer communication and cross-disciplinary training on the following:

4.a Appropriate experimental designs and statistical analyses of postnatal study data for scientists from different disciplines as well as understanding of basic toxicological issues (endocrinology, toxicogenomics, proteomics, epigenetics, etc.) so that research resources are not wasted. It is evident from the literature that special attention needs to be given to assure that studies that examine more than one pup per litter account for this in the analysis of the data. Furthermore, investigators that cross-foster pups should do so properly, retaining both prenatal and postnatal litter information.

4.b Use of reproductive and developmental toxicity data in risk assessment. Promote an understanding of the difference between and no observed effect level and a NOAEL and the need to causally link effects not generally recognized as adverse to clear adverse effects.

5. Maintain an integrated biological approach to study endocrine action (beware of systems biology and the drilling down to molecular levels in vitro as cells and tissues live in complex environments). Integrate EDC latent effects research with other projects on the fetal basis of adult disease.

6. Risk assessment on EDCs

6.a. Develop a scientifically defensible and rational framework for cumulative risk assessments, incorporating the knowledge that dose additivity applies to chemicals that target the same endocrine pathway via very different cellular mechanisms. This redefines what we mean by mode of action and encompasses multiple mechanisms inducing the same phenotypic changes.

6.b. Expanded use of endocrine data in risk assessments not only to support mode of action but also as critical effects.