IDENTIFICATION OF CHEMICALS OF CONCERN WITH RESPECT TO CARCINOGENICITY Vicki L. Dellarco, Ph.D. Office of Pesticide Programs U.S. Environmental Protection Agency 2 nd McKim Workshop on Reducing Data Redundancy in Cancer Assessment May 8-10, 2012 Disclaimer: This presentation does not represent the views and policies of the EPA.

Carcinogenicity Information  Data rich: more reliance on in vivo data requirements  Most pesticides and drugs are tested in two species, two- year carcinogenicity bioassays, generally rats and mice. Genotoxicity data also required  Data limited: more reliance on alternative methods  Most industrial chemicals, pesticide inerts & metabolites/degradates (Q)SAR, Read across/bridging, in vitro (genotoxicity), exposure information  Consider human information and epidemiology when available

Topics  Tumor profiles in rodents & humans  Mode of action approach to evaluate the human relevance of animal tumor  2005 EPA Cancer Assessment Guidelines: Cancer Likelihood Characterization  Directions in Toxicology: 21st Century Initiatives to develop nonanimal (in vitro, in silico) pathway based approaches

Tumor profiles in rodents & humans Topics 2 nd McKim Workshop on Reducing Data Redundancy in Cancer Assessment

Ten Most Prevalent Tumor Sites in Rodents ( Rats (N=564 carcinogens) Mice (N=442 carcinogens) Site No. of Positive Chemicals %Site No. of Positive Chemicals% Liver22240Liver25457 Mammary gland10719Lung12127 Kidney9417Stomach6916 Stomach8816Vascular system6414 Hematopoietic system5710Hematopoietic5412 Lung5810Kidney276 Urinary bladder529Mammary gland225 Nasal cavity / turbinates509Thyroid gland215 Ear / Zymbal’s gland427Urinary bladder123 Esophagus377Uterus

RodentTumor Distribution (219 Pesticides) Liver Lung Thyroid Testes Liver Mouse Rat

Ten Most Prevalent Tumor Sites in Humans (NCI SEER Cancer Statistics Review ) SiteIncidence/100,000 Prostate (male)163 Breast (female)126 Lung & Bronchus79 Colon and rectum59 Urinary bladder37 Skin melanoma25 Lymphoma**24 Corpus uteri (female)23 Kidney & Renal pelvis (male)18 Oral cavity and pharynx 16 ***Non-Hodgkin's lymphoma; 1 age-adjusted to 2000 US population

How Do We Assess Human Health Risks?  Relies heavily on laboratory animal data  Relies on extrapolations, inference methods, safety factors, etc  Animal Biology = Human Biology  Effects found at high animal doses predict effects at environmental levels of exposure  Current animal assays provide adequate coverage for predicting effects on human health including susceptible groups

Mode of action (MoA) analysis approach to evaluate the human relevance of animal tumor response Topic

How Do You Determined the Weight of Evidence (WoE) for Establishing a MoA?  Postulated MoA (theory of the case)  Experimental support for key events Concordance of dose-response relationships Temporal association Strength, consistency and specificity of association of toxicological effect with key events Biological plausibility and coherence  Other possible MoAs  Uncertainties, inconsistencies, & data gaps  Comparison of “Key Events” & relevant biology between animals & humans (qualitative; quantitative) USEPA 2005; IPCS, see

Chemical-Induced Tumorigenesis: Modes of Action  DNA-reactive carcinogens  Chemicals can induce tumors by a variety of MoAs unrelated to DNA damage  Experience from pesticides and/or drugs, e.g., Sustained cytotoxicity & regenerative proliferation Nuclear receptor activation (e.g., PPARa, CAR) & mitogenic proliferation Renal neoplasms in male rats related to alpha-2-u-globulin Urinary bladder neoplasms secondary to mineralization or disruption of normal urinary biochemistry Exaggerated pharmacological effects Immune suppression Hormonal imbalance

2005 EPA Cancer Assessment Guidelines: Cancer Likelihood Characterization Topic

Highlights of 2005 EPA Cancer Guidelines  Hazard assessment emphasizes analysis of all biological information, particularly related to agent’s mode of action  Hazard, dose-response, and exposure characterization  Weight of evidence narrative and standardized descriptors  Major default assumptions are discussed  Framework for judging mode of action information is provided

Weight of Evidence  Narrative: a short summary (one to two pages) that explains an agent's human carcinogenic potential and the conditions that characterize its expression  Descriptors: provide some measure of clarity and consistency in an otherwise free-form narrative  Based on weight of evidence  Are a matter of judgment and cannot be reduced to a formula  Examples are illustrative, NOT a checklist

Weight-of-Evidence Descriptors  Carcinogenic to humans  Likely to be carcinogenic to humans  Suggestive evidence of carcinogenic potential  Inadequate information to assess carcinogenic potential  Not likely to be carcinogenic to humans

Some remarks about the descriptors  Not a check list  For example, when an agent has not been tested in a cancer bioassay, conclusions can still be drawn by scientific inference from toxicokinetic or mode-of- action data  The agent operates through a mode of action for which cancer data are available.  The agent’s effects are caused by a human metabolite for which cancer data are available.

Directions in Toxicology: 21st Century Initiatives to develop non-animal (in vitro, in silico) pathway based approaches Topic

Regulatory Safety Assessment  Meeting Common Needs  A faster, more predictive (relevant) and reliable, and less expensive testing and assessment paradigm that allows focus on chemicals and effects of concern. Move from Empirical to Mechanistic

Enhanced Integrated Approaches to Testing and Assessment 20  Combine existing exposure and toxicity data including information from new technologies (in silico, in vitro and –omics) to:  Formulate hypotheses about the toxicity potential of a chemical or a chemical category.  Target further data needs specific to a chemical or members of a chemical category for a given exposure. Progressive, Tiered-Evaluation Approach: “Integrate, Formulate, Target”

Chemicals Molecular Target Cellular Response Tissue Organ IndividualPopulation Pharmaco- kinetics In vitro studies Biomonitoring Structure Activity Relationships Toxicity Pathways In vivo studies Greater Toxicological UnderstandingGreater Risk Relevance Adverse Outcome Pathway

22 Biologic inputs “Normal” Biological Function Adverse Outcomes (e.g., mortality, Reproductive Impairment) Cell inury, Inability to regulate Adaptive Responses Early cellular changes Exposure Uptake-Delivery to Target Tissues Perturbation Cellular response pathway Molecular initiating event Perturbed cellular response pathway Adverse outcome relevant to risk assessment Toxicity Pathway Adverse Outcome Pathway II.Adverse Outcome Pathways – definition and example Modified From NRC 2007 Pathway-Based Assessment to Predict Adversity

Fit for Purpose  Safety Evaluations  Agricultural chemicals  Antimicrobials and Consumer products  Industrial chemicals  Pesticide inert ingredients  Data Availability/Quality Varies Extensively  Different decisions  Chemical prioritization  Screening level assessment  Quantitative risk assessment  Cumulative risk assessment For Regulatory Purposes Chemical Domain of Applicability Endpoints Duration & Route Decision Context Uncertainty

Level of Confidence (Uncertainties Acceptable?) Decision (Regulatory) Context Human Drug Approval Characterizing toxicity potential of chemicals at Hazardous Waste Site to guide clean-up decision Listing unregulated Drinking Water contaminants to prioritize research and data collection Registration Approval for Agricultural Pesticide Use Comprehensive Data Requirements Data-Limited Situations QualitativeQuantitative Ground Truthing to Apical Toxicity Lower Higher Adverse Outcome Pathway Sorting/Priority Setting for EDSP

21 st Century Methods: Moving Forward 25 Predicted endpoint is defined. Mechanistic interpretation associated with predictions, if possible. Defined chemical domain of applicability for the model. Appropriate measures of goodness of fit, robustness, ability to predict. An unambiguous algorithm. OECD Principles for QSAR Validation: Transparency & Utility for a Specified Application

21 st Century Methods: Moving Forward  Incremental application to decision making.  continuous process of learning and refinement.  In concert with regulatory dialogue.  regulatory frameworks allows the nature of information to evolve in managing chemical risks to ensure effectiveness and efficiency in decision-making process.  understanding the type and degree of uncertainty tolerated in the decision making context will help chart research and incremental application.  Flow from expert peer review and transparency  International harmonization using common frameworks and principles

21 st Century Methods: Moving Forward  Public Outreach  transparency and public participation is mandatory, science necessary but not sufficient  public trust that approach is as good or better than current  incorporation of any new methods would flow from peer review, public participation and transparency Stakeholder support is critical to moving forward