US Environmental Protection Agency
Office of Pesticide Programs
Tools for Integrated Approaches to
Testing and Assessment
June 25, 2009
-------
TOOLS FOR INTEGRATED APPROACHES TO TESTING AND ASSESSMENT
The following tables reflect the Office of Pesticide Program's near (1-5 years) and long term (5-10 years) plans to incorporate new
scientific tools for Integrated Approaches to Testing and Assessment. The priority setting and screening tools listed in Tables 1 and 2
would be applied in the near term within our current assessment practice for making decisions regarding the safety of pesticides for
both humans and wildlife.
The tools listed in Table 1 are designed to make existing animal tests more focused on risk assessment and management needs by
guiding work on an individual chemical or prioritizing work for groups of chemicals.
The key objective of tools listed in Table 2 is to replace an animal test with reliable non-animal methods and to ensure the efficiency
of animal testing by using tiered testing or improving study designs .
Table 3 represents the longer term effort that will transition our current practice into a new risk assessment paradigm using fewer
whole animal tests. This new paradigm relies on understanding how chemicals perturb toxicity pathways.
Table 1: Priority Setting and Screening Computational Predictive Tools.
Table 2: Replacement Tests or Modified Protocols to Traditional Animal Studies
Table 3: Longer Term Tool Development to Support a Paradigm Shift in Testing and Assessment
The tools listed in Tables 1 and 2 will enable a transition into a new paradigm of toxicity testing. Essential to shifting toward a new
paradigm is the building of improved predictive models (i.e, Tables 1 & 2) along with fundamental research that identifies critical
pathways of toxicity and establishes linkages across the different levels of biological organization (chemical interaction with a
molecular target, cellular response, tissue or organ effect and consequent adverse effect on the organism).
6/25/2009
-------
Table 1. Priority Setting and Screening Computational Predictive Tools. In the near term (1-5 years), OPP plans to build
capacity and expand its suite of computational tools to allow more efficient toxicity testing by quickly identifying the likelihood of
potential toxicity effects without conducting the full set of in vivo toxicological studies. These tools largely draw on profiles of
substances with similar physio/chemical properties and biological modes of action.
Goals/Uses/Benefits
Type
Examples of
Tools
Examples of Tools
in Development or
Under Evaluation
QSAR-Based Expert
System for
Predicting
Estrogenic
Activity
Metabolic Simulator
FDA QSAR Models
-expansion to
pesticide chemicals
Example OPP Milestones
• Enhance ability to predict
chemical toxicity by developing
new models and populating
existing models with pesticide
based training sets so that
computational methods are
more useful for pesticides
• Build upon already existing
knowledge for use with data-
limited chemicals, such as the
pesticide inert ingredients,
certain antimicrobial pesticides,
metabolites and environmental
degradates of pesticides as well
as manufacturing process
impurities
• Reduce animal testing by
appropriately directing data
generation toward the most
likely hazards/risk of concern
Models that use existing
knowledge
• QSAR Models
• Expert Systems
• Knowledge Bases
• Read Across from
Analogs/Categories
Existing
• ECOTOX
• ASTER
• ECOSAR
• EPI Suite
• PBT Profiler
• Oncologic
New
• ACTor
• DSSTox
• Metapath
• ToxRefDB
• October 2007 - OPP's Residue of
Concern Knowledgebase
Subcommittee (ROCKS) is
established to provide a systematic
and consistent weight of evidence
approach that fully utilizes
available tools of computational
toxicology to develop hazard
determinations for pesticide
metabolites, residues and
environmental degradates of
concern
• December 2007- EPA hosts
Integrated Approaches to Testing
and Assessment Organization for
Economic Cooperation and
Development (OECD) Workshop
• October 2008 -Letter agreement is
signed between FDA and OPP to
build toxicity databases on
Pharmaceuticals and pesticides to
support better hazard predictions
across different chemical classes
and modes of action
• 2009- February OECD Expert
Consultation and F1FRA
6/25/2009
-------
Goals/Uses/Benefits
Type
Examples of
Tools
Examples of Tools
in Development or
Under Evaluation
Example OPP Milestones
SAP on QSAR-Based
Expert System for Predicting
Estrogenic Activity for food use
inert ingredients and antimicrobial
pesticides
• March 2009 - OPP and Canada's
Pest Management Regulatory
Agency (PMRA) conduct a
validation exercise for
approximately 50 pesticides with
three different QSAR models to
determine whether these models
could predict known pesticide
toxicities.
Models based on
generation of new data:
• Bioactivity
Profiling with in
vitro High Through-
put (HTS) Systems
ToxCast Research
Program
(http ://www. epa. gov/nc
ct/toxcast)
• May 2009 - Analysis of HTS data
on -300 pesticides (ToxCast Data
Summit Meeting:
http ://www. epa. gov/ncct/toxcast/s
ummit.html)
• Spring-Summer 2009- Selection
of additional chemicals (including
inert ingredients and pesticide
active ingredients) for evaluating
the applicability of ToxCast to
predict toxicity potentials
• 2011- SAP meeting on ToxCast
predictive application
• 2012 - ToxCast implementation
6/25/2009
-------
Table 2. Replacement Tests or Modified Protocols to Traditional Animal Studies. These models are intended to replace
animal testing or reduce animal usage in in vivo tests in the near term (1-5 years).
Goals/Uses/Benefits
Type
Examples of Current
Tools
Examples of New
Tools
Example Milestones
• To reduce, refine, and
replace animal testing
for those traditional
animal studies
performed for
purposes of risk
assessment or labeling.
• Non-testing computer-
aided methods to
determine need for a
specific study
• In vitro model to replace
an animal test
• Redesigned animal
study that maximizes the
information gained and
results in a reduction in
the number of animal
studies performed
Standard animal toxicity
guideline tests
QSAR/SAR in lieu of
animal testing
• EPA White Paper on "Use of
Structure-Activity Relationship
(SAR) Information and
Quantitative SAR (QSAR)
Modeling for Fulfilling Data
Requirements for Antimicrobial
Pesticide Chemicals and
Informing EPA's Risk
Management Process" in support
of proposed rule on data
requirements for antimicrobial
pesticides. (Docket: 2008-0110
Document 0045)
http://www.regulations.gov/fdmspub
lic/ContentViewer?objectId=090000
6480665444&disposition=attachmen
t&contentType=pdf
Up and Down Method for
Acute Toxicity (LD50)
Testing (replaces
traditional Acute LD50
Toxicity Test)
Adopted
Draize rabbit eye test
Non-animal approaches
to labeling for eye
irritation hazards, e.g.,
Bovine Corneal Opacity
and Permeability,
May 2009 Interim Pilot using
non-animal assays for labeling
antimicrobial cleaning products
for ocular irritation/hazard (at
http ://www. epa.gov/oppadOO 1)
6/25/2009
-------
EpiOcular, and
Cytosensor
Microphysiometer assays
Guinea Pig Maximization
or Buehler Test
Several in vitro assays
for skin sensitization are
under evaluation or
development that would
replace in vivo testing
Currently under evaluation by
ICCVAM; the assays are
anticipated to be available for use
over the next 1-3 years
2-generation reproductive
toxicity study
Data Requirement:
Reproduction and fertility
effects (Guideline
870.3800)
Redesigned Extended
One-Generation
Reproductive Study
• November 2009: SAP review of
Enhanced Fl Tiered Testing
Approach
• 2010 EPA Test Guideline
• 2012 New Data Requirement in
40 CFR Part 158
Work underway to
design PK studies as part
of standard toxicity
studies to improve dose
selection, dose response
and species
extrapolations
Physiologically Based
Pharmacokinetic (PBPK)
Models to refine dose
selection and animal usage
in toxicity studies
• August 16 - 17, 2007: Assessing
Approaches for the Development
of PBPK Models of Pyrethroid
Pesticides
http://www.epa.gov/scipoly/sap/rn
eetings/2007/081607_mtg.htm
Toxicogenomics data
extracted from traditional
animal studies.
Application of
toxicogenomics to inform
mode of action analysis,
human relevance, and
dose-response.
Spring 2010. OPP/SAP to review
the applicability of
toxicogenomics in pesticide
cancer assessment using
conazoles as an example.
6/25/2009
-------
Table 3. Longer Term (5-10 years) Tool Development to Support a Paradigm Shift in Testing and Assessment.
Goal / Uses/Benefit
Examples of Types of Tools
Develop the means to move, in a scientifically credible and transparent
manner, from a paradigm that requires extensive animal hazard testing and
generation of exposure data, to a paradigm that provides the means to use a
risk-based, hypothesis-driven approach that is based on full use of
computational toxicology tools. These tools must be grounded by
knowledge of toxicity pathways that are perturbed under realistic exposure
conditions
More accurate and focused risk assessments and risk management
Risk assessments based on understanding of mode of action
Substantial reduced reliance on animal testing
HTS and "omics" methods (genomics,
transcriptomics, proteomics,) to inform
mode of action and identification of
toxicity pathways.
Virtual Organ Models (e.g., virtual liver
and embryo)
System biology approaches
New generation of environmental
modeling tools for fate and transport and
human exposure
Well characterized biomarkers of effect
and exposure
6/25/2009
------- |