L. Earl Gray Jr. and Gerald Ankley

1. A Two-Tiered-Testing Decision Tree for Assays in the USEPA-EDSP Screening Battery: Using 15 years of experience to improve screening and testing for endocrine active chemicals L. Earl Gray Jr. and Gerald Ankley This presentation does not necessarily reflect USEPA policy, but rather represents the authors’ current view on the state of the science

2. EDCs – from 1991 to 1996 – Wingspread and Our Stolen Future

3. 1996 – FQPA and SDWA mandates endocrine screening Food Quality Protection Act PUBLIC LAW 104–170—AUG. 3, 1996 110 STAT. 1489Public Law 104–170An Act To amend the Federal Insecticide, Fungicide, and Rodenticide Act and the Federal Food, Drug, and Cosmetic Act, and for other purposes.

4. Page 1532 • ESTROGENIC SUBSTANCES SCREENING PROGRAM.— • ‘ 1) DEVELOPMENT.—Not later than 2 years after the date of enactment of this section, the Administrator shall in consultation with the Secretary of Health and Human Services develop a screening program, using appropriate validated test systems and other scientifically relevant information, to determine whether certain substances may have an effect in humans that is similar to an effect produced by a naturally occurring estrogen, or such other endocrine effect as the Administrator may designate…” • ‘‘2) IMPLEMENTATION.—Not later than 3 years after the date of enactment of this section, after obtaining public comment and review of the screening program described in paragraph • by the scientific advisory panel established under section 25(d) of the Federal Insecticide, Fungicide, and Rodenticide Act or the science advisory board ….” • “SUBSTANCES.—In carrying out the screening program described in paragraph (1), the Administrator— (A) shall provide for the testing of all pesticide chemicals; and • ‘‘(B) may provide for the testing of any other substance that may have an effect that is cumulative to an effect of a pesticide chemical if the Administrator determines • that a substantial population may be exposed to such substance.

5. COLLECTION OF INFORMATION.— ‘‘(A) IN GENERAL.—The Administrator shall issue an order to a registrant of a substance for which testing is required under this subsection, or to a person who manufactures or imports a substance for which testing is required under this subsection, to conduct testing in accordance with the screening program described in paragraph (1), and submit information obtained from the testing to the Administrator, within a reasonable time period that the Administrator determines is sufficient for the generation of the information…” ‘‘(B) PROCEDURES.—To the extent practicable the Administrator shall minimize duplicative testing of the same substance for the same endocrine effect, develop, as appropriate, procedures for fair and equitable sharing of test costs, and develop, as necessary, procedures for handling of confidential business information….” FAILURE OF REGISTRANTS TO SUBMIT INFORMATION.— ‘‘(i) SUSPENSION.—If a registrant of a substance referred to in paragraph (3)(A) fails to comply with an order under subparagraph (A) of this paragraph, the Administrator shall issue a notice of intent to suspend the sale or distribution of the substance by the registrant….”

6. Tier 1 Screening Battery Design - EPA “The EDSP Tier 1 battery was designed to work as a whole with all of the screening assays. The basis for selecting an assay to include in the battery involved two principal aspects: 1. The capacity of an assay to detect estrogen- and androgen-mediated effects by various modes of action including receptor binding (agonist and antagonist) and transcriptional activation, steroidogenesis, and hypothalamic-pituitary-gonadal (HPG) feedback……2. In addition, rodent and amphibian in vivo assays were selected for the proposed battery based on their capacity to detect direct and indirect effects on thyroid function (hypothalamic-pituitary-thyroidal, HPT, feedback). Thus, the robustness of the proposed battery is based on the strengths of each individual assay and their complementary nature within the battery to detect effects on the E, A or T hormonal systems. “

7. Purpose of the Screening Battery • To Detect EAT activity • Not to define dose response or adversity Hence the screening battery does not to include • sensitive in utero or other life stages • All adverse effects of EAT • The most sensitive effects of EAT • T1S needs to include assays with endpoints that are sensitive enough to detect all pesticides and toxic substances that display EAT activity • Does the battery as a whole fail to detect and chemicals with EAT activity - Not that I am aware of

8. Purpose of the testing • Not to Detect EAT activity, but rather • To define dose response and adversity • Hence T2 Testing does need to include • All sensitive life stages • The most sensitive effects of EAT • T2T protocols need to include endpoints that are sensitive enough to be protective for all the adverse effects of pesticides and toxic substances but they do not need to detect all adverse effects produced by disruption of EAT pathways • Do T2T guideline protocols contain the most sensitive effects of EAT? – Not for some modes action

9. The Tier 1 battery’s suite of in vitro and in vivo screening assays includes the following: In vitro assays to detect E and A activity Estrogen receptor (ER) binding – rat uterine cytosol Estrogen receptor - (hERα) transcriptional activation - agonist mode Androgen receptor (AR) binding – rat prostate cytosol Steroidogenesis – Human cell line (H295R) Aromatase – Human recombinant microsomes In vivo to detect EAT activity Uterotrophic (rat) Hershberger (rat) Pubertal female (rat) Pubertal male (rat) Amphibian metamorphosis (frog) Fish short-term reproduction Replace with recombinant protein assays Androgen receptor transcriptional activation assays Estrogen receptor - (hERα) transcriptional activation - antagonist mode Addassays Structure T1S as a Logic Based Decision Tree Strategyusing knowledge gained over the last 15 years on the performance of these assays

10. Tier 1.5 – after T1S but prior to T2 Testing • Objectives of T1.5 – additional in vitro or short-term in vivo assays • To confirm equivocal T1S results or • to explore potential EAT effects in more detail before initiating extensive T2 Testing • For example: An unknown chemical • produces estrogenic effectsin • ER binding and ERTA assays • Uterotrophic assay with sc dosing • Fish assayBut, the chemical does not display estrogenicity in the pubertal female assay • Run the chemical in T1.5 in the uterotrophic assay with oral dosing – if negative, are you really going to T2T with dietary exposure????

11. Implementation of T1S orders and screening – Phase 1 has begunrecommendations forExecution of assays

12. Overt Toxicity is Incompatible with Endocrine Screening !!!!Data exceeding the MTD in the guidelines should be excluded - rejected from the analysis – False Positives Endocrine Disrupted???????

13. Laws SC, Stoker TE, Ferrell JM, Hotchkiss MG, Cooper RL. Effects of altered food intake during pubertal development in male and female wistar rats. Toxicol Sci. 2007 Nov;100(1):194-202.

14. Female Pubertal Assay. Reduced weight gain results in reproductive and other organ weight reductions

15. Female Pubertal Assay. Reduced weight gain results in reduced adrenal weight and reductions in serum thyroid hormones, leptin and glucose levels

16. Male Pubertal Assay. Reduced weight gain up to 20% results in reduced reproductive organ weights without effect on Preputial Separation or serum T

17. Male Pubertal Assay. Reduced weight gain results in organ weight reductions

18. Male Pubertal Assay. Reduced weight gain results in reduced adrenal weight and reductions in serum thyroid hormones, leptin and glucose levels

19. Strategy for Interpretation of T1S battery results • Delete all results from any further consideration conducted at doses that exceed the MTD and caused overt toxicity • Consider “Key” endpoints expected to be affected by either E, A or T disruption as a cluster for consistency across endpoints with a common Mode Of Action (MOA) • However, it is critical to recognize that different responses for SARMS, SERMS, routes of exposure, species differences in sensitivity and the lack of ADME can result in all the “key” endpoints for a MOA may not respond as expected. • Don’t ignore other adverse effects – adrenal weight, gonadal histology, etc

20. How EDSTAC (1998) envisioned interpretation of the results of T1S battery and WoE determinations

22. A Logic Based Decision Tree Strategy for execution of the EDSP screening assays We proposed the use of two in vivo assays as “Gatekeepers” If the “Gatekeepers” were negative for EAT then chemical is placed in the “HOLD” box and the other in vivo and in vitro assays would not need to be executed. If the “Gatekeepers” displayed any positive effects for EAT then specific assays would be executed on a case-by-case basis depending upon the effects seen in the “Gatekeeper” assays Ankley GT, Gray LE. Environ Toxicol Chem. 2013 . Apr;32(5):1084-7. Cross –species conservation of endocrine pathways: a critical analysis ot tier 1 fish and rat screening assays with 12 model chemicals.

23. Detection of Model EDCs by EDSP Tier 1 Assays All pluses are not equal

24. Summary: ER Agonists • EE2 and MXC positive in Uterotrophic (increased uterine weight), female pubertal (advanced vaginal opening) and fish (male VTG induction) assays • BPA negative/low activity in rat assays with oral dosing (female pubertal, Uterotrophic assays), but strong estrogenic positive in Uterotrophic assay (sc route) and fish test (male VTG induction) • Consistent with substantial first-pass (hepatic) BPA metabolism with oral but not sc or inhalation (fish) routes • Responses reflective of HPG functional conservation, but highlight importance of route of exposure (and metabolism) considerations in extrapolation

25. Summary: AR Agonists • TB and MT positive for androgenic activity in Hershberger assay (increased AR-responsive organ weights) and fish (production of male SSCs in females) test; MT also positive in male pubertal assay (advanced puberty [TB not tested]) • MT also positive for estrogenicity in Uterotrophic and fish assays (increased uterine weight; VTG induction) • Consistent with established conversion of MT to M-E2 by aromatase (CYP19) in rats and fish • Cross species conservation of HPG function effectively reflects “paradoxical” (multi-modal) nature of MT

26. Summary: AR Antagonists • FLU, VIN, DDE all produced responses consistent with AR antagonism in two rat assays and the fish test • Hershberger: decreased weight of AR-responsive tissues in animals co-treated with TP • Male pubertal: increased time to preputial separation • Fish: Abnormal ovarian histology (atresia) and decreased expression of SSCs in males (VIN) • FLU and VIN both require metabolism to active AR antagonists (OH-FLU, M3), indicating conserved routes of metabolism in mammals and fish Endpoints quite specific for AR antagonism, interpretation clear Endpoints apical. Could result from many endocrine and non endocrine modes of action. Interpretation not always clear

27. Summary: Steroid Synthesis Inhibitors • KTC positive in both pubertal assays (increased preputial separation time in males; increases in adrenal weight in females) and in the fish test (compensatory increases in GSI of both sexes and proliferation of Leydig cells in males) • FAD, FEN and PRO all positive in fish test (depressed female VTG, E2); results inconsistent in rat Tier 1 assays • FAD delays vaginal opening in female pubertal assay • PRO inhibits AR-sensitive tissue weights in male pubertal assay • FEN, no effect in female pubertal assay, but this pathway is disrupted in multigenerational studies causing adverse low dose effects • Lack of sensitivity of rat Tier 1 assays to some steroidogenesis inhibitors addressed by fish test and H295R assay (which flags KTC, FAD, FEN and PRO all as positives)

28. Summary: Phthalates -PEs • PEs positive in male pubertal assay • Increased preputial separation age in males; decreased SAGs and testis weight • PEs – generally negative for EAT activity in • AR and ER binding, ARTA and ERTA, H295R and AROM • Pubertal Female assay • Fish assay • Frog assay • HTS assays • Lack of sensitivity of all Tier 1 assays to PEs excepts the pubertal male rat assay was the logic for including it in T1S by EDSTAC and as a “Gatekeeper” in our Logic-based Decision tree for T1S assays

29. Tier 1 Screening Gatekeeper Assays Pubertal male - Primary assay for T, A anti-A Fathead minnow Primary assay for SIS, E, A FH and PM Negative + Responses in Gatekeepers Altered T, KT or E2 Altered T4, T3, TSH, thyroid histology Altered T, PPS, Sex Accessory Glands or LH E2, Vitellogenin Male Traits Vitellogenin Triggered T1S Assays ER binding ERTA H295R Aromatase Hershberger Assay and AR Binding ARTA H295R Aromatase Pubertal Female or Oral Uterotrophic Frog Assay

30. Why in vitro assays cannot be “Gatekeepers: • In vitro assays in T1S • Cannot account for ADME • Missing Pathways of EAT action. There are no in vitro assays in T1S for thyroid disruptors, disruptors of hypothalamic pituitary function, and chemical classes like the phthalates that do not interact with nuclear receptors or directly inhibit hormone synthesis • In vitro assay data cannot be integrated at this time to predict the in vivo endocrine effects • In vitro assay data cannot be integrated with other modes of toxicity to predict in vivo toxicity • HTS assays are even more limited • Have not been validated following methods used by EDSP and OECD • Cannot currently meet the EDSP in vitro assay performance standards for sensitivity and reproducibility • Lack assays for steroidogenesis, in addition to thyroid, HPG and phthalates

31. Using Endocrine Toxicity Profiles for specific MOAs to Enhance Tier 2 Testing. using examples of chemicals with estrogenic activity

32. Enhancing Testing on a case-by-case basis using the information from the screening battery • It is evident from the literature that none of the current regulatory multigenerational or one-generation test guidelines include the most sensitive endpoints by chemicals that disrupt estrogen and androgen signaling pathways • These endpoints could be added to standard test guideline studies on a case-by-case basis depending upon the endocrine MOA identified in the T1 Screening Battery. • For example • chemicals that act as ER agonists in vivo via a relevant route of exposure should include an assessment of the mammary gland of the male rat. • Chemicals that display AR agonist activity need to carefully examine the female offspring for retained male sex accessory tissues • Studies of chemicals that produce low incidences of reproductive malformations need to increase the sample size examined for these lesions to have sufficient statistical power. • Studies of chemicals with antithyroid activity need to add assessments of serum T4, T3 and TSH and thyroid gland weight and histology

33. NMDRC and Linear No Threshold Dose ResponseImpact on EDC screening and testingImpact for EDC screening – NONEDo the EDC screening assays fail to detect E or A activity? NOImpact for multigenerational testingEstimation of shape of the dose response curve in the low dose region could be enhanced by using more dose groups For example, keep total N litters in a study as is, increase number of dose groups from 3 to 6 with half as many litters per dose groupExamining more than one animal per litter enhances endpoint sensitivity by increasing the statistical power to detect low dose categorical effects like malformations and histopathological lesions (Blystone et al., 2010; Hotchkiss et al., 2008).

34. Summary • Develop a Logic-based decision tree strategy for T12S assays • Enhance tier 2 testing using results from T1S based upon effects in T1S • Knowledge of the field and understanding of published literature on EDCs is essential to accomplish the above goals

35. Outline of presentation • EDCs – from 1991 to 1996 – Wingspread and Our Stolen Future • 1996 – FQPA and SDWA mandates endocrine screening • 1996-1998 – EDSTAC (the assays, debates over modes of action included) • The final battery – EAT in vivo and in vitro • The next – 15 years and a significant increase in the database on effects of EDCs in EDSP assays and assay validation • Implementation on the first round of chemicals • Cautions on interpretation of Tier 1 Screening data and recommendations for data interpretation of the data. • Recommendations of structuring Tier 1 screening assays on a decision-logic-tree basis with two in vivo assays as the “Gatekeepers” • Description of the value of Tier 1.5 before going to Tier 2 testing. • Using the information from Tier 1 Screening to tailor/enhance Tier 2 testing by adding additional endpoints sensitive to specific modes of endocrine action • What are the endpoints sensitive to disruption that are not specifically included in tier 2 testing? • What is the shape of the dose response curve for EDCs in the low dose range for these sensitive endpoints - where dose it matter: In tier 1 or Tier 2?