Lynne Haber Toxicology Excellence for Risk Assessment Presentation to the CPSC April 8,

Outline of Talk Risk assessment paradigm Overview Exposure assessment Hazard characterization Dose-response assessment Risk characterization Children’s risk issues

Risk Assessment (organizing & analyzing to set priorities & guide management) hazard identification/ characterization dose-response exposure estimation risk characterization social economic public response political Risk Management (decision & action ) 3

NAS 2008 Approach

Risk Assessment/Management Components Hazard characterization: Is this toxic to humans? Dose-response assessment: How toxic is it? Exposure assessment: Who is exposed, how much, how often, and for how long each time? Risk characterization: So what? Risk management: So what will be done about it? Don Barnes, 1993

“The Dose Makes the Poison” Even seemingly safe substances can be toxic at the “right” dose (e.g., water) Organisms respond to toxic substances according to the dose that gets into the body; in toxicology this is called dose-response Increasing the dose or exposure to the substance generally causes more effects Example RfDs: acetone mg/kg-day, zinc mg/kg-day, Aroclor x mg/kg-day

Exposure Assessment What are the sources and routes of exposure? (exposure pathway) What is the magnitude, duration and frequency of the exposure? What are the characteristics of the exposed population? The exposure portion of the paradigm – sources, activity patterns, containment approaches, etc. – provide much of the opportunities for risk management CPSC does nice research to evaluate potential exposure from specific products – e.g., leaching to mimic effects of mouthing or swallowing, wipe tests to mimic handling

Hazard Characterization Considers: Data on toxicity Various endpoints Mechanistic data/Mode of action How a chemical induces an effect Kinetics – what the body does to the chemical Dynamics – what the chemical does to the body Uses weight of evidence approach

Types of Data – Chronic Limits Human data – case reports, epidemiology, controlled studies in limited cases (e.g., air pollutants) Animal data – e.g.: Carcinogenicity Neurotoxicity Reproductive/developmental toxicity Immunotoxicity General system toxicity – e.g., effects on liver, kidney Bioavailability – how much enters the body In vitro studies – mechanistic, “omics”

Types of Data – Acute Limits Data typically much more limited Lethality data (e.g., LD 50 ) commonly available, but are of limited use for identifying safe levels Some research has been done and studies published on extrapolation approaches – e.g., from LC 50 – but involve assumptions and larger uncertainties May have data on some effects – sensitive endpoints typically not evaluated Developmental toxicity still relevant – potential short windows of susceptibility Steepness of dose-response curve is important

Importance of Mode of Action Hazard characterization Determining human relevance (e.g., alpha-2u-globulin related kidney tumors in male rats) Determining conditions of carcinogenicity and other effects (e.g., route-specific at portal of entry; dose-relationship) Dose-response assessment Determining approach for risk characterization/low-dose extrapolation Informing specific aspects of quantitation (e.g., dosimetry, UFs, data needs, childrens’ risk)

Reference Dose (RfD) is... an estimate (with uncertainty spanning perhaps an order of magnitude) of a daily exposure to the human population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime. (Barnes and Dourson, 1988) Non-Cancer Dose Response Assessment…As Defined by EPA

Acceptable Daily Intake (ADI) NOAEL, LOAEL or BMD ADI = UF Where: NOAEL is the No Observed Adverse Effect Level LOAEL is the Lowest Observed Adverse Effect Level BMD is the benchmark dose UF is one or more uncertainty factors needed for data deficiencies 13

Determining the Point of Departure The numerator of the ADI identifies a threshold for toxicity in the experimental species Traditionally, as default, agencies assumed that for most types of effects, there is a dose level below which a response is unlikely Homeostatic, compensation and adaptive mechanisms in the cell protect against toxic effects Critical effect is the first adverse effect or known precursor relevant to humans that occurs as dose increases 14

Areas of Uncertainty Addressed in ADIs and Similar “Safe Doses” – Uncertainty Factors Variability Extrapolating from animals to humans Human variability and sensitive populations Uncertainty Extrapolation from above threshold to below threshold (“LOAEL to NOAEL”) Extrapolation from less than lifetime exposure Consideration of data gaps

16 Dose Response Assessment at CPSC: Non-Carcinogens No-observed adverse effect level (NOAEL) Acceptable Daily Intake (ADI) = NOAEL  10 if human data NOAEL  100 if animal data LOAEL  1000 if NOAEL not established Benchmark dose may also be used

The Continuum Default  Data-Informed Default – 10 x 10 Database - Derived - databases of information, not group or chemical-specific Categorical - applies to categories of substances/species based on their characteristics (BSA correction; RfC - gases/particles) Chemical Specific Adjustment Factors - addressing kinetic or dynamic aspects with chemical specific or compound-related information (e.g., PBPK) Fully Data-Derived – biologically-based dose-response modeling addressing kinetic and dynamic aspects (BBDR)

CPSC - Dose Response Assessment: Carcinogens Linear dose response No threshold Unless there is convincing evidence to the contrary (mode of action) Surface area scaling Humans are 5- to 10-fold more sensitive than rodents

Risk Characterization Integrates exposure data and dose-response (informed by hazard characterization) to obtain risk estimates Provides risk managers with information regarding the probable nature and distribution of health risks Has both quantitative and qualitative components Clearly delineates uncertainty and data gaps

Risk Characterization at CPSC Non-Carcinogens “Hazardous” if exposure > ADI Carcinogens “Hazardous” if cancer risk > That is, the ADI is the dose at risk

21

Available Frameworks ILSI 2003, Workshop to Develop a Framework for Assessing Risks to Children from Exposure to Environmental Agents EPA 2006, A Framework for Assessing Health Risks of Environmental Exposures to Children WHO 2006 (draft), Principles for evaluating health risks in children associated with exposure to chemicals CalEPA 2008, Technical Support Document For the Derivation of Noncancer Reference Exposure Levels

Children’s Differences in Exposure Different exposure scenarios Mouthing behavior – age-dependent Important for child-related products (toys, clothing) Hand-to-mouth activity Near the floor More active More food and water ingested per kg body weight Accounted for in 1-day and 10-day health advisories for water (EPA) Higher breathing rate – may or may not affect internal dose Child –Specific Exposure Factors Handbook. EPA

Where is Children’s Risk Considered in the Paradigm? Specific exposure scenarios As part of human variability – this can be refined based on data, as noted in following slides As part of consideration of overall database – consider whether have identified the critical effect, with particular attention to reproductive and developmental toxicity studies – quantitatively or qualitatively

Differences in Toxicokinetics Toxicokinetics = absorption (entering the body), distribution (where it goes), metabolism (how it is transformed), excretion (how it is removed) There are age-related differences in all of these aspects, but the focus is primarily on absorption/deposition for inhalation exposures, and on metabolism. E.g.: Infants have higher body water Lower levels of key metabolic enzymes Kidney excretion low in first year All of these differences approach adult levels by end of first year or earlier

Aspirin (salicylates) Differences in metabolism between children and adults Children are more susceptible to toxicity of methyl salicylate (oil of wintergreen) Metabolic differences: Children have increased risk of acidosis when overdosed Reye’s syndrome thought to be related to aspirin and age. Children seem to be particularly at risk

Acetaminophen Differences in metabolism between children and adults Different “safe and effective” OTC dose. (Children allowed ~ 30% more per unit body weight) Adults more susceptible to hepatotoxicity when overdosed Children have no significant risk of hepatotoxicity (Increased excretion of sulfate-acetaminophen) Difference in metabolism Children (<12 years)45-55% sulfate, 15-30% glucuronide Adults30% sulfate, 50% glucuronide

Implications of Age-Related Toxicokinetic Differences Interpretation depends on the time period of focus – neonates differ in sensitivity by >3x for some chemicals, but duration of exposure as neonate is short, difference in sensitivity drops rapidly and is small by 1 year Difference in sensitivity for neonates may be critical for acute exposures, need to look at mode of action for chronic interpretation If toxicity is due to cumulative exposure, and average dose is important, increased internal dose to neonate has small impact, due to short duration If chronic toxicity is due to peak exposure, the higher internal dose for neonates may be important

Conclusions Little information is available to refine considerations of toxicodynamic differences, but consideration of mode of action and pathways affected can be useful. Toxicokinetic differences are largest the first year of life. The quantitative implications of those differences depends on the exposure duration of interest and whether peak exposure is important The ratio of mean child:mean adult dose was generally 2 or less, but the ratio can get large for certain combinations of active chemical form and age-related differences in clearance After the first year of life, estimated mean child and mean adult doses are within ~50%.

30

Acute Effects at High Levels Weak and brittle bones “itai – itai disease” Bone become soft, lose mineral density, susceptible to fracture Kidney tubular failure Irreversible Children’s risk? Developing bones may be at higher risk

Chronic Effects of Cadmium Exposure Kidney is the most sensitive target Cadmium accumulates in the kidney, and cumulative exposure across all routes (oral, inhalation, dermal) determines toxicity Urinary excretion is related to total body burden, and can be used to estimate cumulative dose Toxicity measured as a biomarker of effect – protein in urine

Child Susceptibility to Chronic Effects Because of the cumulative nature of the effect, children tend not to be at higher risk while children, but have more time to accumulate cadmium, and so may be at risk later in life There is some evidence that adults exposed to cadmium as children may be more susceptible (for same internal dose) than people exposed only as adults

Cancer Inhalation exposure to cadmium associated with lung cancer in some worker studies and in laboratory animals Several organizations consideration cadmium a known human carcinogen; others say “probable.” Neither animal nor human data are adequate to determine whether cadmium is carcinogenic via oral route Note route difference – mode of action

35

Toxicodynamics: WHO 2006 Key Conclusions Regarding Critical Windows of Development The greatest concern: agents that influence key signalling pathways, affect cell fate, or react with DNA Different effects may result from exposure during early life stages and following adult exposure. Effects shortly after fertilization are hard to identify and evaluate Transient changes in physiology or endocrinology at critical periods of development can result in permanent changes in organ function.

Impact of Adult/Child Differences in Ventilation Rate (Adapted from Clewell et al. 2004) Uptake of vapors depends on ventilation rate only until steady state is achieved At steady state, variation in ventilation rate (e.g., difference between adults and children) does not produce a significant variation in vapor uptake Impact of ventilation rate is different for reactive gases – for those, normalized by surface area