Transcript cg-sra - Society for Risk Analysis

EPA’s cancer risk
assessment guidelines:
General overview
Jim Cogliano, Ph.D.
United States Environmental Protection Agency*
Office of Research and Development
National Center for Environmental Assessment
Washington, D.C.
*The views expressed in this presentation are those of the author and do not necessarily
reflect the views or policies of the U.S. Environmental Protection Agency.
Topics to be covered
• Key features of the new guidelines.
– Use of defaults.
– Emphasis on mode of action.
– Weight-of-evidence narrative and descriptors.
– Dose-response assessment: two-step approach and
choice of linear or nonlinear extrapolation.
– Cancer risks from early-life exposure.
• Anticipated schedule and opportunities for interaction.
Risk assessment
HAZARD
ASSESSMENT
Can the agent cause cancer?
Is this relevant to humans?
Who may be more sensitive?
DOSE-RESPONSE
ASSESSMENT
Estimate a dose-response curve
Account for high-to-low-dose,
animal-to-human, route-toroute, and other differences
RISK
CHARACTERIZATION
EXPOSURE
ASSESSMENT
How do people come in contact
with the agent?
How much are they exposed to?
Integrate HAZARD, DOSERESPONSE, and EXPOSURE
Describe the assessment’s
strengths, limitations, and research
needs
Why change the guidelines?
• New kinds of studies (mechanistic studies) are giving
insight into how a chemical causes cancer
• The 1986 guidelines do not discuss how to evaluate
mechanistic studies
• For some carcinogens, the mechanistic studies indicate
that the dose-response curve is likely to be nonlinear at
low doses
• The 1986 guidelines do not discuss how to evaluate
nonlinear dose-response relationships
Key features of the new guidelines
• Emphasis on analyzing data before invoking defaults.
• Emphasis on mode of action throughout the guidelines.
• Weight-of-evidence narrative replaces the “A-B-C-D-E”
classification scheme; descriptors can be route-specific.
• Two-step dose-response process separates (1) modeling
the observed data from (2) extrapolation to lower doses.
• Linear and nonlinear extrapolations are considered.
• Differential risks to children are addressed.
The 2002 draft emphasizes analysis
of data before use of defaults
Analyze the available data
Is there too much uncertainty or
is critical information lacking?
Y
Invoke a
default option*
N
* “The primary goal of EPA actions is public health
protection, accordingly, as an agency policy, the
defaults used in the absence of scientific data to the
contrary should be health protective (SAB 1999).”
Uses of mode of action in the 2002
draft guidelines
• Assess the relevance of laboratory animal results to
human environmental exposures
• Identify sensitive populations and lifestages
• Provide insight into whether the dose-response curve is
likely to be linear or nonlinear at low doses
• Quantify the relative sensitivity of laboratory animals and
human populations
Framework for evaluating support for
an hypothesized mode of action
a. Is the hypothesized mode of action sufficiently supported
in the test animals?
– Examination loosely patterned after the “Hill criteria.”
b. Is the hypothesized mode of action relevant to humans?
– Considers all populations and lifestages.
– Anticipated low human exposure levels are not used
to conclude that a mode of action is not relevant.
c. Which populations or lifestages can be particularly
susceptible to the hypothesized mode of action?
– Question is both qualitative and quantitative.
– Quantitative differences are flagged for use in the
dose-response assessment.
The “Hill criteria”
Used by epidemiologists to infer whether an observed
association may be a causal association.
a.
b.
c.
d.
e.
f.
g.
h.
i.
Consistency of the observed association.
Strength of the observed association.
Specificity of the observed association.
Temporal relationship of the observed association.
Biological gradient (exposure-response relationship).
Biologic plausibility.
Coherence.
Experimental evidence.
Analogy.
Weight-of-evidence narrative in the
2002 draft guidelines
• Conclusions about human carcinogenic potential,
including a weight-of-evidence descriptor.
• Conditions of carcinogenicity:
– Route, magnitude, and duration of exposure.
– Susceptible populations and lifestages.
• Summary of key evidence supporting these conclusions.
• Summary of key default assumptions.
• Summary of potential modes of action.
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
• Multiple descriptors can be used if the cancer hazard is
route-dependent or dose-dependent.
• 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, for example:
– 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.
Two-step dose-response assessment
in the 2002 draft guidelines
Tumor
incidence
STEP 1. Model
the observed data
down to a point
of departure
xx %
?
STEP 2.
Extrapolate
to lower
doses
POD
Dose (mg/kg-d)
Characterizing the point of
departure: the POD narrative
a. Nature of the response.
b. Level of the response.
c. Nature of the study population.
d. Slope of the dose-response curve at the POD.
e. Relationship of the POD with other cancers.
f. Extent of the overall cancer database.
Choosing linear or nonlinear
extrapolation
• Linear extrapolation is appropriate:
– When the agent is mutagenic or acts through another
mode of action expected to be linear at low doses, or
– Human exposure or body burden is high and near
doses associated with key precursor events.
• Linear extrapolation is also used as a default when
the data do not establish the mode of action.
• A slope factor is developed in these cases.
• Nonlinear extrapolation is appropriate
– When there is no evidence of linearity, and
– There is sufficient information to support a mode of
action that is nonlinear at low doses.
Linear extrapolation under the 2002
draft guidelines
Tumor
incidence
STEP 1. Model
the observed data
down to a point
of departure
xx %
STEP 2.
Extrapolate
to lower
doses
POD
Dose (mg/kg-d)
Nonlinear extrapolation under the
2002 draft guidelines
Tumor
incidence
xx %
STEP 2.
POD
Extrapolate
to lower
exposures of interest
doses
STEP 1. Model
the observed data
down to a point
of departure
Dose (mg/kg-d)
Dose-response characterization in
the 2002 draft guidelines
•
•
•
•
•
Recommended estimates (for example, slope factors).
Summary of the data supporting these estimates.
Summary of the modeling approaches used.
The POD narrative.
Identification of susceptible populations or lifestages and
quantification of their differential susceptibility.
• Strengths and limitations of the dose-response
assessment, highlighting:
– Significant issues
– Alternative approaches considered equally plausible.
– How these issues were resolved.
Children’s risk in the 2002 draft
guidelines
• “Children’s risk” can mean different things to different
people:
– Effects manifest during childhood.
– Early-life exposures that can contribute to effects at
any time later in life.
• In the cancer guidelines, we are interested in both.
Some reasons why cancer risks can
differ following early-life exposure
• Differences in capacity to metabolize and clear chemicals.
• More frequent cell division during development:
– Enhanced expression of mutations due to reduced
time for repair of DNA lesions.
– Clonal expansion of cells with unrepaired DNA
damage.
• Immune system that is not fully functional.
• Hormonal systems that operate at different levels.
• Potential for developmental abnormalities to result in a
predisposition to carcinogenic effects later in life.
Review of the data pertinent to
cancer risks from early-life exposure
• Comparison of cancer bioassays using mature animals
with those using developing animals.
• EPA’s vinyl chloride assessment on IRIS.
• Other chemicals with studies of cancer following early-life
exposure.
• Site-specific human cancer risks following exposure to
radiation at different ages.
EPA is developing new approaches
for assessing children’s cancer risks
• Examine the data pertinent to cancer risks following
early-life exposures.
• Develop approaches that are:
– Consistent with the state of the science.
– Health-protective when critical information is absent
or uncertain.
• Allow these approaches to be updated when there is new
information or new understanding.
Anticipated schedule and
opportunities for interaction
By end of 2002:
• EPA expects to release two new drafts:
– Guidelines for carcinogen risk assessment.
– Supplemental guidance on assessing risks to children.
• At this time, EPA will ask for:
– Executive-branch agency review.
– Public comment (60 days).
– One-on-one outreach meetings with stakeholders.
Early in 2002:
• EPA expects to begin addressing the comments received.
• EPA will obtain peer review of the scientific basis of the
supplemental guidance on assessing risks to children.
Thank you.