Health-Related%20Components%20of%20DSL%20Categorization%20under%20CEPA%201999%20%20Exposure%20and%20Hazard%20Tools%20Presented%20by:%20%20Jesse%20Ng%20Existing%20Substances%20Division%20Health%20Canada%20Oct%2020,%202005

1
Health-Related Components of DSL Categorization
under CEPA 1999 Exposure and Hazard
ToolsPresented by Jesse NgExisting
Substances DivisionHealth CanadaOct 20, 2005
2
Outline

• The Canadian Environmental Protection Act
• Health Canadas mandate
• Categorization
• Principles and objectives
• Maximal List
• Exposure Tools
• Hazard Tools
• Input from stakeholders
• Screening Assessments
• Key messages

3
Existing Substances under the Canadian
Environmental Protection Act (CEPA)

• CEPA is administered jointly by Environment
  Canada and Health Canada
• CEPA 1999 extended our mandate from Priority
  Substances to Categorization of the approximately
  23,000 existing substances on the Domestic
  Substances List (DSL) by September 2006
• The DSL was created for the purpose of defining a
  new substance under CEPA
• Includes substances grandfathered under the
  legislation
• Substances in use between January 1, 1984 and
  December 31, 1986
• Organics (50), organic metal salts,
  organometallics, inorganics, polymers and
  substances of unknown or variable composition,
  complex reaction products biological materials
  (UVCBs)

4
Health Canadas Mandate on Existing Substances

• Address both exposure and effects to set
  priorities for risk assessment and management
  under CEPA
• Source characterizations to inform risk
  management
• Information Gathering/Industrial Surveys
• Publicly accountable transparent process and
  content, peer input, consultation and review,
  documented outcome

5
Categorization human health

• Need to consider
• Greatest potential for exposure (GPE) all DSL
  substances
• Inherently Toxic to humans (IThuman) subset
  of substances
• Those that are P or B but not inherently toxic
  to environmental organisms (ITeco)
• Challenges
• Categorization must be completed by September
  2006
• Consistency with Priority Substances outcomes for
  high hazard

6
CEPA Existing Substances Program
CATEGORIZATION of the Domestic Substances List
(DSL) (First Phase) (n23,000)
Decisions of Other Jurisdictions
Public Nominations
STAGE 1
STAGE 2
Greatest Potential for Human Exposure
Substances that are Persistent or Bioaccumulative
Inherently Toxic to Humans
Inherently Toxic to non-Human Organisms
STAGE 3
SCREENING ASSESSMENT (Second Phase)
No further action under this program
Risk Management
CEPA-Toxic
IN-DEPTH ASSESSMENT Priority Substances List
(Third Phase)
No further action under this program
CEPA-Toxic
Risk Management
STAGE N-1
STAGE N
7
Categorization Objectives

• Set priorities for data generation and health
  assessment for all Existing Substances
• Health protective approach, conservative in the
  absence of information
• Complex program architecture requires multiple
  stages of increasing complexity which address all
  groups of compounds concomitantly
• First stages simple/pragmatic to address all
  substances, based on limited information for each
  or many
• Simple tools
• Subsequent stages must be discriminating to set
  true priorities for further work
• Complex tools
• Avoid continuing bias to focus on data-rich
  Existing Substances

8
Identifying Highest Priorities for Human Health
Approach

• Initial application of simple, discriminating
  tool on exposure to address all 23,000 substances
  to prioritize Greatest Potential for Exposure
  (GPE), Intermediate Potential for Exposure
  (IPE) Lowest Potential for Exposure (LPE)
• Draws on information submitted in compilation of
  the Domestic Substances List
• Application of simple, discriminating tool to
  address hazard for all 23,000 substances
• Draws on work completed internationally
• Priority-based application of more complex tools
  to additionally refine prioritize

9
INTEGRATED FRAMEWORK
DOMESTIC SUBSTANCES LIST
HEALTH CANADA
ENVIRONMENT CANADA
Application of Simple Tools
Substances that are Persistent and/or
Bioaccumulative According to the Regulations
Highest
DSL Substances Identified as Hazardous to Human
Health
Organic Substances that are Persistent and/or
Bioaccumulative and Not Inherently Toxic to
Non-human Organisms
DSL Substances Ranked According to Potential For
Exposure
Lowest
Substances that are Persistent and/or
Bioaccumulative and Inherently Toxic to
Non-human Organisms
HC Maximal List
Application of Complex Tools
EC Substances Identified for Screening Assessment
Substances Prioritized Identified For Full
Screening Health Assessment
No Further Action (Not 64c toxic)
10
The Maximal List
High 576
301 LPE, High Hazard
275 GPE or IPE High Hazard
Moderate 989
480 GPE
121 IPE, P or B
388 IPE, P or B unknown
183 Low Hazard
Low 331
148 other
11
What Do the Groups on the Maximal List Mean?

• High likelihood of remaining for further work
  beyond 2006
• Subset of 301 for risk management (LPE high
  hazard)
• Moderate likelihood of remaining as health
  priorities beyond 2006
• Information will help here
• UVCBs, polymers, wide range use substances
  delineated as priorities
• Low likelihood of remaining for further work
  beyond 2006
• Low hazard
• Substances already addressed CEPA

12
DSL TOOLS - HEALTH

• Exposure
• SimET (Relative ranking of all DSL substances
  based on submitters (S),quantity (Q) and expert
  ranked use (ERU)
• ComET (Quantitative plausible maximum
  age-specific estimates of environmental and
  consumer exposure for individuals based on use
  scenario (sentinel products), phys/chem
  properties bioavailability)

• Hazard High (H) or Low (L)
• SimHaz (identification of high or low hazard
  compounds by various agencies based on weight of
  evidence)
• ComHaz (Hierarchical approach for multiple
  endpoints data sources (e.g., QSAR) including
  weight of evidence

• Hazard Quantification (Previously
  Exposure-Response)
• HazQ (measures of exposure-response developed
  (where possible) on the basis of measured or
  predicted carcinogenic potency, reference values
  or effect levels

13
Simple Exposure Tool - SimET

• SimET is a relative ranking tool by which we have
  binned all substances on the DSL
• Considers potential for environmental and
  consumer exposure
• Based on three different lines of evidence,
  derived from the limited information provided for
  all substances on the DSL
• quantity (estimated annual quantity of use, Q),
• number of submitters, S
• use (sum of normalized expert ranked use codes,
  U), reflecting two workshops
• Industrial sector and functional use codes

14
Criteria for Greatest, Intermediate and Lowest
Potential for Exposure (GPE, IPE LPE)
Quantity (kg/year) Number of Submitters Sum of the Expert Ranked Use Code Indices
GPE gt 100 000 Top 10 Top 10
IPE gt 10 000 n.a. Top 30
LPE All All All
15
The Complex Exposure Tool (ComET)

• Provides quantitative plausible maximum estimates
  of exposure of individuals in the general
  population by age group for consumer (near-field)
  multimedia environmental (far-field) exposure
• Far-field exposure
• Based on concentrations in environmental media
  estimated from fugacity modelling
• Near-field exposure
• Frequency and duration of product use
• Based on Sentinel product scenarios
• Exposure for all age groups to be addressed

16
Overview of ComET
17
Principles in Developing ComET

• Transparency in the approach and assumptions
• Uncertainties and data gaps identified
• Defensibility, Consistency and Inclusiveness
• Drawing maximally on the documented work of
  others by building on existing scenarios
• Data call in to stakeholders
• Peer input, consultation and review
• Outreach to other jurisdictions
• Fit for purpose
• Conservative and protective assumptions for
  priority setting, but adaptable to enable
  incorporation of more refined models
• Readily useable and applicable to all chemicals
  irrespective of data available
• Drawing maximally on generic information

18
ComET Far-field

• Exposure from DSL substances in the environment
• Extension of existing fugacity models (e.g.,
  ChemCAN) to estimate concentrations of substances
  in environmental media (Mackay model)
• Physical/Chemical properties
• Emissions/Releases
• Distribution of substances into relevant media,
  e.g., air, water, soil, sediment
• Fate, e.g., drinking water, foodweb
• Generic unit world model that can be scaled and
  modified for further refinement
• Applied to substances for which little or no
  empirical property data are available and
  emission rates are known only approximately

19
The Far-field Model
Substance
Phys-Chem properties and use information (Substanc
e Profile)
Calculate Unit Emissions
Transformation and fate
Migration into media for uptake
Scale unit concentrations using actual
emissions, production quantities and use
information
Ambient Concentrations
Integrate with Near-field Component
20
Far-field - Output
Route of exposure Estimated intake (µg/kg-bw per day) of (name of substance from DSL) by various age groups Estimated intake (µg/kg-bw per day) of (name of substance from DSL) by various age groups Estimated intake (µg/kg-bw per day) of (name of substance from DSL) by various age groups Estimated intake (µg/kg-bw per day) of (name of substance from DSL) by various age groups Estimated intake (µg/kg-bw per day) of (name of substance from DSL) by various age groups Estimated intake (µg/kg-bw per day) of (name of substance from DSL) by various age groups Estimated intake (µg/kg-bw per day) of (name of substance from DSL) by various age groups
Route of exposure 06 months 06 months 0.54 years 511 years 1219 years 2059 years 60 years
Route of exposure formula fed not formula fed 0.54 years 511 years 1219 years 2059 years 60 years
Ambient air
Indoor air
Drinking water
Food and beverages
Soil
Total intake
21
ComET Near-field

• Exposure from DSL substances in consumer products
• Selection of sentinel products
• Identification of use of a substance for a
  specific function in a product (e.g., surfactant
  in paint, solvent in paint, pigment in paint)
• Physical/Chemical properties
• Bioavailability
• Sentinel Product Scenario
• Contains elements of exposure, i.e., exposure
  factors
• Maximum proportion generically used for a
  specific function in a product (e.g., of
  surfactant in paint, etc.)
• Frequency and duration of product use
• Amount transferred during use
• Age-specific personal factors
• Designed to provide a reasonable worst-case
  estimate of exposure

22
Sentinel Product (SP)

• A sentinel product is a specific type of consumer
  product with a defined composition and use that
  yields the highest exposure to an individual for
  one of its component substances as compared to
  other consumer products containing that substance
• Sentinel products are selected from broader
  classes, e.g. personal care products
• A specific substance is then matched to one or
  more sentinel product(s) based on generic
  information about its use pattern
• e.g. for acetone, possible SPs are
• nail preparations
• acrylic paints
• There may be more than one SP for a given
  substance

23
Considerations for selection of Sentinel Products

• Product categories generally accepted to
  represent high exposure potentials
• Household cleaning products
• Soaps and detergents
• Cosmetics and personal care products
• Food additives
• Fabric treatments
• Paints and coatings
• Adhesives and sealants
• Hobby and craft products
• Automotive care and maintenance products
• Lubricants
• Fuels and solvents
• Lawn and garden care products

24
Scenarios for Sentinel Products

• Routes of exposure (oral, dermal, inhalation)
  determine how scenarios are developed
• Information gathering
• In-house contracts, surveys, e.g., CEPA sect. 71
• MSDS, e.g., CCOHS, NIOSH
• Public sources, e.g., Scorecard, Household
  Product Database, etc.
• Industry input, e.g., Soap and Detergent
  Association (SDA)
• Appropriate algorithms are selected from an
  exposure matrix including ComET and other
  publicly available models (e.g., ConsExpo, ECETOC
  TRA, CEM, etc.)
• Appropriate conservative exposure factors are
  used to populate the algorithms (e.g., Versar,
  SDA, etc.)
• ComET contains 146 different Sentinel Products
  scenarios, each of which contains one or more
  exposure route(s)

25
Scenario Algorithm - Example

• Inhalation
• ED WF x A x ET x IR x EF
• BW x RV x AT
• where
• ED estimated dose per event (mg/kg-bw perday)
• WF weight fraction of substance in product
• A amount of product used per event (mg)
• ET exposure time (i.e., duration of exposure)
  (h)
• EF Exposure frequency (unitless)
• IR inhalation rate (m3/h)
• BW body weight (kg-bw)
• RV room volume (m3)
• AT Averaging time (day)

Age-specific variables
26
ComET output Near-field exposure

• When a substance occurs in more than one product
  or is described by more than one use code, ComET
  will provide an estimate of either
• Sum of doses or highest dose
• Estimate values for any of the different types of
  exposure
• Acute, sub chronic, or chronic
• Any of the six age groups
• Any of the three route specific exposures or the
  total dose

27
ComET Next Steps

• Completion of taxonomy of sentinel products and
  algorithms for sentinel products
• Solicit input/information/comment on the
  architecture, and data to populate for decision
  making
• Availability of habit and use surveys
• Example SDA Document exposure and risk screening
  methods for consumer product ingredients
• Peer review of taxonomy, algorithms and default
  values
• Need for adequate documentation as a basis for
  default parameters in the algorithms and
  selection of sentinel products
• Public availability of habit and use survey
  information
• Consistency with well documented sources
• Process peer input, consultation, review

28
Exposure Tools

• Phase
• Categorization
• Screening Assessments
• In-depth Assessments -
• Priority Substances

• Tool
• Simple Exposure Tool
• Complex Exposure Tool
• Consumer exposure scenarios/algorithms
• Consumer Exposure Models
• WPEM (wall paint exposure model),
• CEM (consumer exposure module, E-Fast),
• ConsExpo

29
DSL TOOLS - HEALTH

• Exposure
• SimET (Relative ranking of all DSL substances
  based on submitters (S),quantity (Q) and expert
  ranked use (ERU)
• ComET (Quantitative plausible maximum
  age-specific estimates of environmental and
  consumer exposure for individuals based on use
  scenario (sentinel products), phys/chem
  properties bioavailability)

• Hazard High (H) or Low (L)
• SimHaz (identification of high or low hazard
  compounds by various agencies based on weight of
  evidence)
• ComHaz (Hierarchical approach for multiple
  endpoints data sources (e.g., QSAR) including
  weight of evidence

• Hazard Quantification (Previously
  Exposure-Response)
• HazQ (measures of exposure-response developed
  (where possible) on the basis of measured or
  predicted carcinogenic potency, reference values
  or effect levels

30
SimHaz Tool

• Applied to entire DSL
• Defines high or low hazard from
  classifications/assessments of other agencies
  based on weight of evidence
• Appropriate assessments selected based on
  comprehensiveness of review, peer review process,
  etc.

31
SimHaz Tool

• Endpoints chosen based on general population
  concerns
• High Hazard Lists/Endpoints
• Cancer (IARC, EU, HC, US EPA etc.)
• Genotoxicity (EU)
• Developmental Toxicity (EU)
• Reproductive Toxicity (EU)
• Low Hazard Lists
• PMRA 4a/US EPA
• OECD Low Concern
• Respiratory sensitization endpoint dropped from
  SimHaz as more relevant to occupational exposure

32
SimHaz ToolStrengths and Limitations

• Strengths
• Efficient
• Takes advantage of critical review of others
• Consistency
• Assessments/classifications internationally
• Limitations
• Bias towards data-rich substances

33
ComHaz Tool

• Hierarchical approach for multiple endpoints
  data sources, including limited weight of
  evidence, for identifying compounds for further
  consideration.
• - Qualitative and/or quantitative criteria
  developed for various endpoints.
• - Conservative so that confidence is high that
  substances that are not considered priorities for
  further consideration based on any of the
  criteria are non hazardous.
• - For qualitative endpoints, weight of evidence
  is assessed, where possible.
• Currently, confidence in predictive tools only
  for cancer/genotoxicity

34
Complex Hazard (ComHaz) Tool
35
ComHaz Tool

• Sources of Information
• Comprehensive literature searching (electronic
  hardcopy resources)
• Reviews or secondary accounts of toxicological or
  epidemiological studies
• Original Toxicological and Epidemiological
  Studies
• (Quantitative) Structure Activity Relationship
  (QSAR)Models (TOPKAT, CASETOX)
• Chemical structures of concern, Structure
  Activity Relationship (SAR) models (DEREK),
  surrogate/analogue approaches (Leadscope)

36
ComHaz Tool

• Qualitative criteria
• Cancer
• Genotoxicity
• Developmental toxicity
• Quantitative criteria
• Regulatory/reference values
• Developmental toxicity
• Reproductive toxicity
• Long term toxicity
• Shorter term toxicity
• Acute toxicity

37
ComHaz Tool Criteria
Endpoint Information Source Criteria
Cancer Data or (Q)SAR Weight of evidence
Genotoxicity Data or (Q)SAR Weight of evidence
Regulatory/Reference Value International National Assessments Ref Value 0.1 mg/kg bw/day 0.4 mg/m³
Developmental Toxicity Data NO(A)EL 90 mg/kg bw/day NO(A)EC 270 mg/m³
Developmental Toxicity (Q)SAR Positive Prediction
Reproductive Toxicity Data NO(A)EL 10 mg/kg bw/day NO(A)EC 30 mg/m³
Longer Term Toxicity Data or (Q)SAR (where appropriate) NO(A)EL 10 mg/kg bw/day NO(A)EC 30 mg/m³
Short Term Toxicity Data NO(A)EL 30 mg/kg bw/day NO(A)EC 90 mg/m³
Acute Toxicity Data or (Q)SAR (where appropriate) LD50 500 mg/kg bw LC50 1500 mg/m³
38
ComHaz ToolPreliminary Weight of Evidence (WoE)
Framework

• Why Cancer/Genotoxicity?
• ComHaz endpoints for which capture rate is
  highest
• Qualitative ComHaz criteria are very conservative
  (i.e., first hit)
• Need to increase discrimination to identify
  priorities for further consideration
• Confidence in (Q)SAR greatest for these endpoints
• Larger more diverse training sets (e.g., simple
  screening assays such as Ames test)
• Potential for combining relevant endpoints
• Relevance to specific modes of action
• Genotoxic carcinogenicity is critical endpoint
  for more in-depth assessments (i.e.,
  screening/PSL)

39
Preliminary WoE FrameworkDevelopment Process

• Draft approach developed
• Acquired operational experience through
  consideration of individual compounds
• Continued to revise approach based on this
  experience as well as internal and external
  consultation
• Internal consultation with genotox specialists
• External peer consultation (www.tera.org)

40
Preliminary WoE FrameworkPrinciples/Approach

• Separate consideration of endpoints
  Carcinogenicity Genotoxicity
• Separate consideration of lines of evidence
• Empirical data, QSAR, SAR
• For (Q)SAR models, output is weighted based on
  predictive power of both the assays and
  validation results for similar compounds.
• Equivocal data/inconclusive predictions noted,
  but not weighted
• Call for a line of evidence based on
  consideration of ratio of positives/negatives and
  degree of confidence
• Degree of confidence based on consistency between
  data and predictions

41
Preliminary WoE Framework (Q)SAR Models
42
Preliminary WoE Framework Outcomes
CHEMICAL X From Moderate Group of Maximal List
WoE Cancer/Genetox High Potential for Genotoxic
Carcinogenicity?
YES
Positive carcinogenicity study or QSAR prediction
NO
YES
Positive genotoxicity study or QSAR prediction?
NO
Meets criteria for remaining quantitative
endpoints in Hierarchical Approach?
NO
YES
NO
YES
Next Step Hazard Quantification
Tool (Exposure-Response Characterization)
SET ASIDE
43
ComHaz ToolProcess Development/Testing

• Considerable internal operational experience
• External testing for consistency of output based
  on search strategy/approach
• External peer review of internal/external
  consistency of critical aspects of approach
• Internal QA/QC
• Expert consultation
• e.g., genotoxicity, WoE

44
ComHaz ToolStrengths and Limitations

• Strengths
• Health protective
• Comprehensive
• High confidence in set asides
• No bias towards data rich substances
• Designed for high throughput
• Takes advantage of critical reviews of others
• Significant contribution of QSAR component to
  international priority setting
• External input, consultation, peer review
• Limitations
• Resource intensive

45
Hazard Quantification Tool (Previously called
Exposure-Response Tool)

• Developed from a Toxicity Profile
• For compounds that are ComHaz IN
• All toxicological endpoints considered (i.e.,
  carcinogenicity, developmental, reproductive,
  acute, etc.)
• For each endpoint, the Quantified Hazard
  (carcinogenic potential, NOEL/LOEL, etc) is
  determined for various durations of exposure
• Available data on pharmacokinetics, mode of
  action and species specificity are also
  considered
• Informs as to what type of Screening Model will
  be proposed for compound (ie. Various screening
  assessment models are being drafted in some
  cases, input from Exposure Tools will be
  required).
• Toxicity Profile and Tool still in development

46
Hazard Tools

• Phase
• Categorization
• Screening Assessments
• In-depth Assessments -
• Priority Substances

• Tool
• Simple Hazard Tool (SimHaz)
• high and low hazard
• Complex Hazard Tool
• first stage
• QSAR/SAR WoE (if needed)
• Exposure-Response (Hazard Quantification)Tool

47
Input from Stakeholders

• 60-day comment period on the Proposed Integrated
  Framework for the Health-Related Components of
  DSL categorization ended August 30, 2005.
• Data requested on DSL compounds, especially those
  on the Maximal List, Deadline for submission was
  Sept. 16, 2005.
• Responses received from Industry
• e.g., ACC, ATOFINA, BASF, CPMA, Degussa, Nova
  Chemicals
• Input from Environmental Non-Governmental
  Organizations

48
Screening Assessments

• Need to assess more compounds more quickly
• No legislated deadlines, however high
  expectations
• Draw on international/assessments to extent
  possible and considerable collective experience
  in HC and limited external peer review
• Consistent with principles of in-depth PSL
  assessments

49
Full Focused Screening Assessments

• Decisions based on consideration of
• Nature of critical effect
• Margin between critical effect level and upper
  bounding estimate of exposure
• Adequacy of margin to account for uncertainties
  in database
• Possible outcomes
• Not toxic under CEPA 1999
• Further in-depth assessment required
• Toxic under CEPA 1999

50
Screening Assessments - Status

• Public comments received on PBDEs PFOS
• Screening Health Assessments to be released on
  our Listserv
• Quinoline
• MBMBP MBOCA
• 1,2-Dibromoethane 1,1-Dichloroethene
• Biphenyl Ethylbenzene
• DNOC Hexachloroethane

51
(No Transcript)
52
Key Messages

• Framework developed to identify true priorities
  from a human health perspective.
• takes into account both exposure and hazard.
• Series of simple and complex tools developed.
• exposure assessment and hazard identification.
• Tools applied in identification and
  prioritization of substances for assessment.
• also as part of the health risk assessment
  process itself.
• Innovative and ensures efficient assessment to
  meet CEPA 1999 mandate.

53
More Information?

• Health Canada Existing Substances Division
  Website http//www.hc-sc.gc.ca/ewh-semt/contami
  nants/existsub/index_e.html
• Health Canada Maximal List and Integrated
  Proposal Framework http//www.hc-sc.gc.ca/ewh-s
  emt/contaminants/existsub/framework-cadre_e.html
• Health Canada Existing Substances Mailing List
  http//www.hc-sc.gc.ca/ewh-semt/contaminants/exis
  tsub/mail-avis_e.html
• CEPA Registry
• http//www.ec.gc.ca/CEPARegistry/default.cfm