&EPA
US EPA Office of Research and Development
United States
Environmental Protection
Agency
Office of Research and
Development
Washington DC 20460
EPA/600/R-00/105
December 2000
Summary of the
Workshop on Information
Needs to Address
Children's Cancer Risk
-------
-------
EPA/600/R-00/105
December 2000
Summary of the Workshop on
Information Needs to Address
Children's Cancer Risk
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC 20460
Printed on Recycled Paper
-------
NOTICE
This document has been reviewed in accordance with U.S. Environmental Protection
Agency (EPA) policy and approved for publication. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
This report was prepared by Eastern Research Group, Inc. (ERG), an EPA contractor
(Contract No. 68-C-98-148, Work Assignment No. 2000-01) as a general record of discussions
during the Workshop on Information Needs to Address Children's Cancer Risk. This workshop
was co-sponsored by EPA's National Center for Environmental Assessment and National
Institutes of Health's National Institute of Environmental Health Sciences. As requested by
EPA, this report captures the main points and highlights of discussions held during plenary
sessions. The report is not a complete record of all details discussed nor does it embellish,
interpret, or enlarge upon matters that were incomplete or unclear. Statements represent the
individual views of each workshop participant; none of the statements represent analyses by or
positions of the NIEHS or the EPA.
11
-------
CONTENTS
Page
INTRODUCTION j
THURSDAY, MARCH 30 2
WELCOME AND CHARGE TO WORKSHOP PARTICIPANTS
William'Farland and Michael Firestone J......... 2
SENSITIVITY OF CHILDREN TO ENVIRONMENTAL TOXICANTS
Lynn Goldman 4
Discussion 6
CHILDREN'S ENVIRONMENTAL HEALTH
Lynn Goldman , g
Discussion _ 9
FRIDAY, MARCH 31 13
EXPOSURE OF CHILDREN TO ENVIRONMENTAL TOXICANTS
Philip Landrigan '...'.' 13
TOPIC 1: CURRENT AND PROPOSED APPROACHES TO ASSESSING
CHILDREN'S CANCER RISK
William Farland .....'. 16
Discussion .i 20
Facilitated Discussion ...;........ 21
Discussion 24
TOPIC 2: ENHANCED USE OF TEST DATA RELATED TO CHILDREN'S CANCER
RISK
Rochelle Tyl 26
Facilitated Discussion 27
General Discussion 28
m
-------
TOPIC 3: FUTURE DIRECTIONS FOR TOXICOLOGY TESTING TO ADDRESS
CHILDREN'S CANCER RISK
RethaNewbold 32
Facilitated Discussion 33
Discussion 35
TOPIC 4: EPIDEMIOLOGIC/MOLECULAR EPIDEMIOLOGY INFORMATION TO
ADDRESS CHILDREN'S CANCER RISK
Leslie Robison , • • 36
Facilitated Discussion 38
General Discussion 40
WORKSHOP SUMMARY
George Lucier
42
Discussion 43
APPENDIX A: WORKSHOP PARTICIPANTS A-l
APPENDIXB: LIST OF OBSERVERS B-l
APPENDIX C: MEETING AGENDA C-l
APPENDIXD: CHARGE TO PARTICIPANTS • D-l
APPENDIX E: OVERHEADS USED IN THE PRESENTATIONS E-l
APPENDIX F: LIST OF BACKGROUND MATERIALS PROVIDED TO
PARTICIPANTS PRIOR TO THE MEETING F-l
IV
-------
INTRODUCTION
On March 30-31, 2000, the U.S. Environmental Protection Agency (EPA)'s Office of
Research and Development and the National Institute of Environmental Health Sciences
(NIEHS) cosponsored a workshop entitled "Information Needs to Address Children's Cancer
Risk." The workshop focused on a discussion of children's cancer risk assessment and related
data needs to address issues that were raised during public review of The Agency's 1999 Draft
Revised Guidelines for Carcinogen Risk. These issues include:
• Characterizing the ideal data set to adequately address children's cancer risk.
• Proposed approaches to using available data in the absence of the ideal data set.
The background for discussions at the Workshop is the reality that chemical-specific data
are often lacking to specifically address children's cancer risk from environmental chemical
exposures. Consequently, the assessment of children's risk is currently addressed by evaluations
of traditional bioassays in mature animals, comparative biochemistry and physiology between
adult and developing animals and humans, and public-health-protective default positions in the
absence of child-specific data. The Workshop focused on four topics areas:
• Topic 1: Current and Proposed Approaches to Assessing Children's Cancer Risk.
• Topic 2: Enhanced Use of Test Data Related to Children's Cancer Risk.
• Topic 3: Future Directions for Toxicology Testing to Address Children's Cancer Risk.
• Topic 4: Epidemiological/Molecular Epidemiology Information to Address Children's
Cancer Risk.
-------
Summary ofEPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
The cosponsors invited the participation of leaders in the area of human health testing,
research, and assessment who represented the pediatric, lexicological, and risk assessment
communities. The invited participants addressed not only the induction of childhood cancer, but
also increased risks of cancer during adulthood as a consequence of childhood exposure.
Observers participated in the discussions of issues specific to topic areas and contributed
comments during periods of general discussion.
This report summarizes the Workshop discussions. Appendix A lists the Workshop
participants, and Appendix B provides a list of observers. The meeting agenda and charge to
participants can be found in Appendices C and D, respectively. Appendix E contains copies of the
overheads used in the presentations. Appendix F lists the background materials provided to
participants prior to the meeting.
THURSDAY, MARCH 30
WELCOME AND CHARGE TO WORKSHOP PARTICIPANTS
William Farland and Michael Firestone
William Farland, Director of EPA's National Center for Environmental Assessment, and
George Lucier of NffiHS welcomed the participants and observers on behalf of the sponsoring
Agencies.
Michael Firestone of EPA's Office of Children's Health Protection noted that the intent of
EPA and NffiHS in sponsoring the Workshop was not to seek recommendations or reach
consensus decisions. The main purpose, he said, was to obtain individual views and perspectives
on children's cancer risk assessment and related data needs, and to address issues that have arisen
during review of EPA's Draft Revised Guidelines for Carcinogen Risk Assessment. He said
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31, 2000
Workshop participants should focus on how discussions might have an impact on the ongoing
effort to revise EPA's Cancer Guidelines. He listed the specific issues that would be the focus of
discussion during the Workshop (see "Charge to Workshop Participants" in Appendix D and
"Charge to the Children's Cancer Workshop Participants" in Appendix E):
Characterizing the content of the ideal data set to adequately address children's cancer
risk, with a focus on data needed for assessing the impact of childhood (including in
utero) exposures to carcinogens and the issues related to hazard identification and dose-
response analyses.
Addressing not only induction of childhood cancer, but also increased risks of cancer
during adulthood resulting from'childhood exposure.
Considering how current bioassay testing protocols might be redesigned to better answer
questions related to children's cancer risk and what additional types of data might be
brought to bear on children's cancer risk assessment.
Defining what are the elements of a "cogent biological rationale," as presented in the draft
revised cancer guidelines, for addressing modes-of-action for children's cancer.
Answering whether and how a "cogent" rationale that is sufficiently health-protective of
children can be made based on the kinds of data that are typically collected by and
available to Federal and State health science agencies at the present time.
Defining what additional data, such as cancer mode-of-action and comparative
pharmacokinetics and pharmacodynamics in adults and children, might be useful in
developing a "cogent" rationale.
Addressing whether the assessment of children's risks, as it is currently conducted by
evaluations of traditional bioassays in mature animals using sensitive responders, is
sufficiently public-health-protective in the absence of child-specific data.
In discussing these points, participants noted that the Workshop's purpose was not to
describe idealized research protocols that might be developed to provide all data necessary to
characterize children's cancer risks, and urged that discussions remain focused on the questions
of how best to use available methods and data to address children's cancer risks. An observer
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
noted that the last issue—whether current use of traditional bioassays in mature animals is
sufficiently protective in the absence of child-specific data—might imply a simple "yes" or "no"
answer, and urged that the Workshop go beyond that answer. Abe Tobia asked whether the
Workshop would be involved in looking at design of future studies, noting that it would require a
significant effort. Lynn Goldman replied that the Workshop's charge allows discussion of new
study designs, but emphasized the need to consider current issues as expressed in the draft
Guidelines. George Lucier encouraged Workshop participants not to get bogged down with a
great deal of detail and specificity when addressing future needs. He also asked them to
remember, during the discussions, that the revised Guidelines should be able to stand the test of
time and adequately capture the need for new approaches and strategies to be used in toxicology
testing as it relates to childhood cancer. Abe Tobia repeated his view that the Workshop should
focus on what is currently being done and potentially relevant to the current Guidelines. William
Farland noted that the Workshop schedule included an opportunity to discuss possible directions
for future research. He suggested that specific changes in protocol should be addressed by a
separate panel or Workshop in the future.
SENSITIVITY OF CHILDREN TO ENVIRONMENTAL TOXICANTS
Lynn Goldman
Dr. Goldman noted that the Workshop faced a challenging task in addressing childhood
cancer and its potential causes as well as children's exposures to carcinogens, which are two
separate but important issues in terms of risk assessment. She said Workshop participants should
be mindful of the fact that EPA and other government agencies represented at the Workshop
make decisions each day with respect to cancer risks and that the goal of the Workshop was to
make positive contributions to those decisions. She noted that the purpose of the Workshop was
not to specifically evaluate EPA's Cancer Risk Assessment Guidelines or to replace other
mechanisms for review of the Guidelines. Dr. Goldman presented an overview of issues related
to childhood cancer:
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31, 2000
• They have a low rate of occurrence and there is uncertainty about trends in occurrence.
• Childhood cancers are limited to a few unique types that are found in children but not
adults and tend to reflect fetal derivation of cells and prenatal exposures.
• Most childhood cancers have short latency periods; many are characterized by known
genetic and familial associations.
• There is a high probability of genetic/environmental interactions in children's
susceptibilities to cancer.
Dr. Goldman noted that childhood cancer mortality has been decreasing but that there
were increases in the rates of acute lymphocytic leukemia and brain cancers among children
1980s. Some research suggests that childhood is a time of greater susceptibility to cancer, Dr.
Goldman said. The possibility that children and the developing fetus face risks not seen in adults
is supported by examples such as in utero exposure to DBS during a specific period of fetal
development and the subsequent-occurrence of vaginal cancer and diseases that resemble birth
defects. This suggests a hormonally driven process that changes cell differentiation. Other
examples of childhood-specific risks are incidences of radiation-induced cancers that have a
short latency and suggest increased risk during developmental periods of rapid cell division.
Exposure to tobacco during periods of rapid cell division may also explain observed relationships
between age of smoking initiation and lifetime risk for lung cancer and a persistence of risk after
people stop smoking. Although only a small percentage of cancers are due solely or in part to
environmental exposures, these cancers may account for 5-25 percent of annual cancer deaths,
and therefore represent a large public health burden. Dr. Goldman suggested that, in considering
the adequacy of the rodent bioassay model, the Workshop's discussions of childhood
susceptibility should focus on:
Genetic susceptibilities, including inherited predisposition and polymorphisms that result
in pharmacokinetics that affect dose.
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
• Reduced latency that results from rapid tissue growth during childhood.
• The persistence of effects due to mutations or alterations of cell differentiation.
• Nutritional factors.
Hormonal factors, including the influence of exogenous hormones such as DBS.
Dr. Goldman asked participants and observers to respond to these points and to add other
issues that would be relevant to the Workshop's goals of considering how the EPA Guidelines
can properly address children's cancer risks.
Discussion
Rochelle Tyl said the Workshop should consider the repair capacity of a young organism
compared with an older more developed organism, the differences in metabolism between
prenatal or perinatal animals and adult animals, and clearance. These issues may reveal reasons
why the young may be more susceptible, Dr. Tyl said.
Paul Foster recommended consideration of the developmental "window" during which
exposure occurs. Chris Portier said that consideration of cancers resulting from viral exposures
also should be considered. Dr. Portier and Dr. Goldman both noted that the interaction of
multiple factors is an important consideration but one that would be very difficult to examine in
bioassays. David Wallinga suggest adding consideration of immature immune systems and the
protective factors such as the the patency of the blood-brain barrier in immature animals. Lauren
Zeise noted that certain exposure factors should be considered; she cited as an example the
increased exposure to contaminants in drinking water of a child being fed infant formula.
-------
Summary of-EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31, 2000
Frederica Perera said that racial, ethnic, or cultural variability may play a role in
susceptibility but cannot be modeled using rodent bioassays. Joseph DeGeorge observed that
possible genetic predispositions and racial or ethnic variabilities may play as large a role in adults
as in children, and might therefore be beyond the scope of the Workshop. Dr. Perera and Dr.
Goldman responded that genetic predisposition may play an important role in cell growth and
differentiation and therefore might be particularly important during periods in life when there is
rapid cell growth. George Lucier said there would be a mushrooming of information about the
relationship between genetic predispositions and diseases that are easily detectable, such as
childhood cancers. How to use that information in childhood cancer risk assessments is going to
be very difficult, he said. Dr. DeGeorge repeated his observation that genetic predispositions and
racial or cultural factors do not represent defining factors between the effects of fetal or
children's exposures and adult exposures, and are a bigger issue than the Workshop's focus on
childhood cancer risks. Mark Miller noted that there are genes that may be associated
carcinogenicity or susceptibility in childhood but are not associated with adult cancers. Chris
Portier said that genetic predisposition is an important area to explore for differences between
adults and children. For example, genes that "turn on" during a particular stage of development
may point out windows of opportunity. Polymorphisms in those genes coupled with exposure at a
certain time could have a serious effect. Because these genes tend to be selected out of the
population, it is very difficult to gather information without specifically looking for it, Dr. Portier
said.
Michael Thun said that an underlying theme in the discussion seemed to be the question of
the conditions under which a study in rodents can give the wrong answer to questions about
childhood cancer. For example, he noted, a study in animals may not show a problem but under a
special circumstance such as nutritional deficiency or the presence or absence of a polymorphism
there may be a problem. The number of possible permutations of conditions that would need to
be studied is huge, and it will be a long time until there is a bioassay that will tell whether there is
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31.2000
a problem in all subjects. EPA now factors in a protective margin—the upper 95 bound—to cope
with this problem, but from the point of view of a biologist, Dr. Thun said, all bioassays provide
incomplete information.
Lauren Zeise suggested that another issue to consider is timing. Exposure early in life has
more time to interact with other exposures to cause cancer, but timing is not now addressed in
Guideline default procedures, she said.
Leslie Robison noted that children's cancers tend to be very specific types of cancers, and
said that extrapolating from data acquired in animal models may not have anything to do with the
induction of the unique spectrum of cancers that occur in children.
Abe Tobia said that 90-day animal assays may not reveal problems but do not allow for a
pathological continuum that leads to some childhood cancers and, from that perspective, short
term studies may yield false negative data. A very important issue, he said, is the need to conduct
a long-term definitive study that detects these cancers and can be used to determine whether that
relates back to childhood. Rochelle Tyl said that the key problem with 90-day studies and chronic
studies is that exposure starts in animals that have gone through puberty and are essentially
adults. This misses the most vulnerable stages of development for some cancers, and even 2-year
or 3-year bioassays that begin exposure at 6 to 8 weeks will not detect cancers resulting from in
utero or lactational exposure.
CHILDREN'S ENVIRONMENTAL HEALTH
Lynn Goldman
It is crucial to recognize that children are not little adults and are exposed in ways that
have no parallel in adult life. For example, breast-feeding is an exposure route only for infants.
Moreover, a child's metabolism may be more or less capable than an adult's of breaking down,
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
inactivating, or activating toxic substances. The rapid growth and development of organ systems
that takes place during childhood increases the vulnerability of children, who also have more
years of future life in which diseases with long latency periods might develop.
Children's exposure to dioxin is more than two orders of magnitude greater during their
first years of life, when they are breast fed, than later in life, Dr. Goldman pointed out. Intake
rates for water, food, and air, per kilogram of body weight, are greater for children than for
adults, and some routes of exposure are different in children. Because infants, toddlers, and
preschool-age children spend much of their time on the floor or on the ground and use their
hands and mouths to explore these environments, they are exposed in different ways to different
contaminants than are adults. In addition, children's diets frequently focus on certain foods that
are relatively uncommon in adult diets, she noted.
Discussion
George Lucier asked what types of information are now available that indicate the relative
magnitude of children's body burdens of toxics compared with those of adults. Dr. Goldman
replied that there is very little available. William Farland said that the EPA was initiating studies
of very small populations as a first step in measuring national human exposures, and is
participating in planning stages of a longitudinal birth cohort study that will provide more
detailed information on exposure and body burden. Chris Portier said that measurement of body
burdens would provide more relevant information than measurement of exposure and intake. Dr.
Goldman said that there are very few animal studies that provide information relating body
burden in mothers to body burden hi the fetus. Joseph DeGeorge noted that intake rates as well as
metabolic elimination rates change rapidly throughout childhood. Dr. Goldman noted that animal
studies may not be able to accurately model these changes in humans. Dr. DeGeorge said that the
Food and Drug Administration had conducted surveys of literature to obtain information on
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
organ development and profiles of metabolism as it changes through childhood. The data are
difficult to find and, he added, it is very difficult to develop a parallel between adult animals and
adult humans, and even more difficult to find juvenile animal models that represent juvenile
humans. Dr. Farland suggested that pharmacokinetic modeling may offer a better understanding
of dose in both humans and animals. Refining these studies, he said, will improve the ability to
understand the effects of target doses on target tissues. This would eliminate the sometimes
misleading reliance on measuring exposure and intake levels.
f rederica Perera noted that the multiple or repeated exposures lead to complex
interactions that are not well understood, and that these interactions may have different effects in
children than in adults. David: Wallinga said that, unlike adults, children have a fairly predictable
set of exposures through types of food or specific medications, but these predictable mixtures
have not been considered in risk assessment. Lauren Zeise suggested that studying body burdens
of compounds that act by similar mechanisms may be more valuable than concentrating on the
body burden of a single compound, particularly when determining whether the observed dose-
response should be considered in a linear or nonlinear way.
Joseph DeGeorge noted that existing data acquired through neonatal rodent assays
demonstrates that juvenile animals are more susceptible than adults to carcinogens. There is no
need to develop new tests to detect differential sensitivity. What is needed, he said, is an
understanding of why there is greater sensitivity and how it applies to humans. George Lucier
noted that for a few effects, such as breast cancer, information exists to show that animal data can
be applied to human risk. Using lessons from these few models that have been well explained
would help develop other mechanistic animal models that can be applied to humans. Dr.
DeGeorge noted that more than 90 chemicals have been tested in juvenile animals, and a
neonatal mouse assay, which can identify tumor effects within a year, is now being studied as an
alternative to the two-year adult bioassay. Dr. Portier noted that much of the published data is
10
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
from studies that rely on one or two doses and does not tell very much about the curvature of the
dose-response. This is important, he said, because there may be chemicals that cause adult cancer
through a very nonlinear mechanism but have a linear mechanism in infants. Dr. Goldman
observed that it is important to fully explore the issue of susceptibility versus exposure, because
differences in susceptibility would result in different dose-response curves in children and adults
with the same exposures.
Dr. Portier noted that the same mechanism of action can have different effects at different
ages or. stages of development. For example, he said, a carcinogen may be activated by
metabolism but is then mediated by an organism's repair capacity. If that capacity is low in the
child and high in the adult, there might be linearity in the child and nonlinearity in the adult, even
though the same mechanism is involved.
Chris Wilkinson said the Workshop should recognize that the EPA's current draft
represents a good set of cancer risk assessment guidelines that should not be further delayed by
extensive discussion of specific children's cancer issues. These issues are very important, he
said, but should not become a barrier to finalizing the Guidelines. He suggested that the
Workshop's goal should be to identify four or five major factors that could be incorporated into
the Guidelines and move forward. Dr. Farland noted that the Workshop has been charged
specifically to address childhood sensitivity and to assure that the guidance put forward is public-
health-protective. Abraham Tobia said the Workshop should focus on identifying a few topics
that can be fully explored. He said the Guidelines represent a solid framework and should be
implemented, and that fuller discussion of other issues will come later. It is important that the
regulated community understand what kind of information it is expected to generate.
David Goldsmith said that susceptibility must be looked at as a result of interactive
effects. He also noted that it is necessary to understand the role of changes, in immunologic
11
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
competence as a child matures. Sam Kacew said that the role of lactation and breast feeding,
which had been mentioned as means of exposure, should also be considered in terms of
development of the immune system. He suggested that the Workshop also consider other
nutritional factors, both protective and harmful. Paul Foster said that fetal dosimetry and
lactational transfers are critical measures that should be developed to provide first-hand practical
information on how much of a chemical crosses the placenta and how much is transferred in
milk.
Jeanette Wiltse said that information about mode of action from a 2-year bioassay or from
a 2-generation study is not relevant when exposure begins in utero. She suggested that
Guidelines should encourage studies on mode of action in the very young animal, which is not
part of the standard protocol. Dr. Tobia said that the regulated community wants to provide
information that will be used in the risk assessment process. Sometimes that information is
outside the required data but may help make a case for a different way of modeling a carcinogen.
But, he added, unless the information is used in decision-making there is no benefit to doing the
work necessary to gather it. Dr. Farland said that relevant information may be available in data
that are routinely collected for endpoints other than children's cancer. Dr. Portier said that it is
important to determine what kind of information is necessary to understand the mode-of-action
issue in children, and that type of information cannot be derived from current data on adult
animals. Philip Landrigan said that it is simply not possible to extrapolate from the adult
experience to predict what is going on in the neonate or the infant. He urged that the Workshop
focus on the role of the Guidelines as a means to protect human health. He said that this goal is
best achieved by assessing the risks in children, who are the most vulnerable segment of the
population.
12
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
FRIDAY, MARCH 31
EXPOSURE OF CHILDREN TO ENVIRONMENTAL TOXICANTS
Philip Landrigan
Workshop chair Philip Landrigan opened the Workshop's second session by observing
that the explicit goal of the Guidelines should be prevention of disease, not detailed
understanding of mechanisms of action. Risk assessment should be considered in a public health
context, he said, and he noted that two centuries of medical advances had dramatically reduced
the incidence and mortality of infectious diseases in the U.S. During the past 50 years the number
of synthetic chemicals entering the environment has increased enormously, and few of these
chemicals have been subjected to toxicity testing. To discuss detailed mechanisms of action for a
few chemicals while basic toxicity data are lacking for many chemicals is putting the cart before
the horse, he said.
Patterns of disease in children are changing in ways that are not well understood, Dr.
Landrigan said, noting that asthma, childhood cancers, congenital urinary tract defects, and
testicular cancer in young men have increased steadily since the early 1970s. Ten years ago the
National Academy of Sciences (NAS) Committee on Pesticides and Children was charged to
answer three questions that are directly relevant to the Workshop's purpose, Dr. Landrigan said.
The questions are:
• Are children more heavily exposed than adults?
• Are children more susceptible to toxicity than adults?
Do current laws and decisions protect children?
13
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
Childhood exposure to carcinogens is vastly different and greater by orders of magnitude
than adult exposure. Moreover, Dr. Landrigan noted, children live and play on the floor and often
put their hands in their mouths, and their exposures come from a wider variety of environmental
sources than do adults'. Children are also more vulnerable to environmental toxicants, as
indicated by such examples as children's increased risk of cancer following exposure to
nitrosamines and vinyl chloride, decreased ability to detoxify organophosphates, increased
susceptibility to lead and alcohol (fetal alcohol syndrome), and the relationship of DBS and
adenocarcinoma of the vagina. Such data led the NAS Committee to note that "children are not
little adults" and to conclude that:
Children's exposures to pesticides are greater pound-for-pound than those of adults.
Children are less well able than adults to detoxify most pesticides.
Children's developing organ systems are highly vulnerable to pesticides.
Children have more years of future life in which to develop chronic disease triggered by
early exposure.
Dr. Landrigan emphasized that the last point is important to the Workshop's discussion
because, although childhood cancers are relatively rare, exposures during childhood increase the
risk of adult cancer. Dr. Landrigan cited the NAS committee conclusion that "compared to late-
in-life exposures, exposures to pesticides in early life can lead to effects that are expressed only
after long latency periods have elapsed. Such effects include cancer, neurodevelopmental
impairment and immune dysfunction." The NAS committee concluded that traditional risk
assessment does not reflect the complexity of children's exposures to carcinogens, is limited to
study of too few chemicals, and largely is based on exposures of adult animals.
Commenting on earlier Workshop discussion suggesting that consideration of children's
cancer risks might add a complicating layer of complexity to risk assessment, Dr. Landrigan said
14
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
that children's risk should be the core of the risk assessment Guidelines. Risk assessment in a
public health context has as its goal the protection of the most vulnerable, Dr. Landrigan said,
and therefore must be based on risks to children. Dr. Landrigan noted that the NAS committee
made general and specific recommendations that were incorporated into the 1996 Food Quality
Protection Act and led to EPA's 1996 declaration that children's health is a specific focus of the
Agency's environmental health plan.
Dr. Landrigan referred to a recent journal article (Faustman, et al.) in which the authors
discuss- consideration of children's susceptibility in an overall framework for human risk
assessment and say "An important public health challenge has been the need to protect children's
health. To accomplish this goal, the scientific community needs scientifically based child-
specific risk assessment methods." That comment, Dr. Landrigan said, should set the stage for
the Workshop discussions. Dr. Landrigan said that defaults and safety factors have become a
major component of pesticide regulation, and are applied when it is determined through research
that infants and children are more vulnerable than adults to a compound or, more commonly,
when there no child-specific data are available. He said defaults are not sufficiently emphasized
in the Carcinogen Risk Assessment Guidelines and suggested that there be explicit mention of
defaults in the Guidelines. In closing, Dr. Landrigan said that despite the work already done to
draft the Guidelines, they should be fundamentally rewritten as a concise document that clearly
states goals and objectives, cites previous work, describes methods, and concludes with a
discussion and references. Other work done to date would be included as an appendix, he
proposed.
15
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
TOPIC 1: CURRENT AND PROPOSED APPROACHES TO ASSESSING CHILDREN'S
CANCER RISK
William Farland
Dr. Farland said the EPA is a public health agency with its principal focus on disease
prevention. To this end, he said, the Agency's draft Guidelines have been developed to protect
the most vulnerable populations and most sensitive individuals. This includes an explicit
consideration of children and their unique vulnerabilities, Dr. Farland noted. He reviewed the
development of Cancer Risk Assessment Guidelines and noted that they serve not only to guide
risk management but to identify research needs and to advance the science of risk assessment,
particularly as it might be applied to children and other vulnerable populations. As a result, he
said, cancer risk assessment is an iterative process and the Guidelines are the product of
continuous dialogue and reevaluation driven by new data and models. He noted that the
Guidelines are the result of interagency colloquia, peer consultation and review, three reviews by
EPA's Science Advisory Board, multiple interagency reviews, and public comment. He said the
Agency Risk Assessment Guidelines are:
Statements of Agency policy regarding principles, general approaches, preferences, and
default assumptions that will be applied in Agency risk assessment.
• Not a cookbook.
• Not a regulation.
The first Agency carcinogen Guidelines were issued in 1976, and new Guidelines based
on the "state of the science'-' were issued in 1986. The 1986 Guidelines were flexible, Dr. Farland
said, but provided little guidance on when or how to depart from default assumptions and
therefore did not provide much incentive to collect better data. Moreover, they did not include
specific consideration of children. These shortcomings led the Agency to initiate the revision of
the Guidelines. The new directions for risk assessment guidelines:
16
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
• Emphasize full characterization.
• Expand the role of mode of action information and, therefore, biomarkers.
• Use all information to design dose-response approach.
• Incorporate a two-step dose-response assessment.
The two-step dose-response assessment first considers information within the range of
observation and then explicitly considers moving from these observations into the range of
inference to make, in some cases, decisions that are not testable. The Draft Carcinogen Risk
Assessment Guidelines reflect a mode-of-action analysis based on physical, chemical, and
biological information rather than a detailed mechanism-of-action analysis that may delay action
because there would never be complete information. Risk assessment has evolved from hazard
identification that relies on traditional toxicologic testing to hazard characterization through
evaluation of mechanisms and biologically based models ranging from new strains of rodents to
mathematical models, Dr. Farland said. Mode-of-action considerations involve asking:
• How does the chemical produce its effect?
• Are there mechanistic data that support this hypothesis?
• Have other mechanistic hypotheses been considered and rejected?
Mode-of-action data is used in dose-response assessment to:
• Construct a biologically-based or case-specific model.
• Link the dose-response curve for precursor effect to dose-response curve for tumor effect.
• Use dose-response for other effects in lieu of that for tumor effect if it is judged to be a
better measure of potential risk.
17
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
Inform assessment of possible dose-response in range of extrapolation.
In the range of extrapolation, Dr. Farland said, the dose-response assessment is:
Linear if:
• DNA-reactive or other evidence supports linearity.
Not DNA-reactive but there are insufficient data to characterize a non-linear mode of
action.
Non-linear if:
not DNA reactive or otherwise linear, and sufficient data exist to characterize a non-linear
mode of action.
Both if:
There is differing activity at different sites.
Linear and non-linear approaches are needed to explain complex activity.
Non-linear includes a margin of exposure approach that is new to the Guidelines, Dr.
Farland said. This is an evaluation of how close the available human or animal data are to the
exposure of interest. It allows for a judgment as to whether or not the increment of exposure is
large enough to give regulators confidence that they are being public-health-protective.
The linear approach is public-health-conservative because, by using an upper bound on
risk, it allows the Agency to project several orders of magnitude from observed data without
making adjustments to account for human variability, Dr. Farland said. This is in agreement with
the National Research Council's suggestion that pharmacokinetic models or scaling adjustments
be applied to account for species differences in toxicokinetics, differences in exposure rate, or the
magnitude of exposure in a population being considered. Low-dose extrapolation is conducted at
18
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
the point of departure—the lower 95 percent confidence limit on the lowest dose associated with
tumor response—determined under standard conditions on test rodents considered to be stressed,
not average. The straight-line extrapolation achieves risk estimates similar to those derived by
the procedures described in the 1986 guidelines, and overestimates risk at low doses. The linear
approach assures that risk to the population is not Underestimated and thereby protects public
health, Dr. Farland said.
Generalized models are not able to account for differences in risk due to human
variability, Dr. Farland noted. Therefore a margin of exposure analysis is used when a nonlinear
default is supported. If no agent-specific data suggest a differential response in children, the
human variability factor is applied with adjustment to account for dose in children, but with no
other additional factors to protect children, he said. The proposed Guidelines' approach to
children's risk incorporates:
• Potential differences in exposure, dose, and response between children and adults.
• A case-by-case approach based on weight of evidence.
• Default science policy positions and procedures to be used in the absence of data.
The Guidelines call for separate evaluation when data suggest increased sensitivity to
exposures that occur early in life and include an illustration of how these data can be applied to
calculate both adult and childhood-specific unit risk estimates. In addition, Dr. Farland said, this
approach provides a lifetime risk estimate that considers, both independently and additatively,
increased childhood risk as well as effects in adults due to early-in-life exposure. Because slope
factors and unit risk for lifetime exposure are based on adult data, the Guidelines adjust adult unit
risk to account for differences in dose between adults and children, Dr. Farland said. These
adjustments involve:
19
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
Default procedures for adult-to-child risk adjustments based on differences in dose:
Oral dose factor—no adjustment is proposed.
Inhalation unit risk (gases)—adjustment based on body weight and breathing rate.
Drinking water unit risk—adjustment based on body weight and drinking water
rate.
• Determining guidance for inhaled particles and dermal exposure.
Asking whether these default procedures are appropriate and incorporating new data.
In considering dose-response in children, if a postulated mode-of-action is supported for
adults but not for children, a linear low-dose default will be applied as a default for the general
population, including children. This approach accounts for the possibility of increased risk to
children while possibly overestimating adult risk, thereby providing public-health-protective
estimates based on possible effects in the most sensitive population, Dr. Farland said. When there
is no available information on mode of action in children, or when there is no cogent biological
rationale that supports the assumption that mode of action in children is the same as in adults, the
postulated mode of action is not considered applicable to children.
Dr. Farland said the Agency hopes to publish the new Guidelines early in 2001, and will
include a shorter supplementary guidance focused on assessment of children's risk.
Discussion
George Lucier asked how much flexibility the Guidelines would allow in order to
accommodate factors such as differences between the ways in which children and adults are
exposed or interspecies variations such as the 100-fold difference in half-life of dioxin in humans
compared with rodents. The lack of information on such issues is a significant problem facing
20
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
implementation of the Guidelines, he said. Dr. Farland noted that the Guidelines must calculate
human equivalent dose from animal models. Individual cases such as dioxin would have to be
considered separately and explicitly, he said, but the standard human equivalent dose approach
includes body weight to the three-quarters power as a scaling factor based on metabolic rate.
Exposures coming from different sources are accounted for, in part, by the Guidelines'
consideration of background exposure when calculating exposure relative to the dose-response
curve, Dr. Farland said.
Lynn Goldman said the Guidelines should clearly describe the mode of action default for
children when data exist only for adults. And, she noted, children are part of the general
population, and referring to them as a "subpopulation" carries the risk of trivializing the issue of
childhood risks, which effect all humans. Dr. Farland said that children's risk assessment begins
by making an argument for a mode of action and then asking whether applying that mode of
action will be protective of the most sensitive individuals in the population. Unless there are data
to suggest that the mode of action applies to children or a cogent biological argument can be
made to suggest that it applies, the Guidelines assume it does not and the linear default is made.
Facilitated Discussion
Abraham Tobia: One of the Guidelines' central points is the default assumption and the
movement from linear to weight of evidence or the cogent argument. The Guidelines allow the
regulated community to generate information that addresses the cogent argument and move away
from the linear default. Dr. Tobia said that industry has begun to look at the cogent argument
issue and to generate data by looking at the young without ignoring the older population. He said
there is a need for more pharmacokinetic and toxicokinetic data that address questions of
saturation and differential sensitivity between the young and old, and he urged that the
21
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
Guidelines allow industry to be flexible in developing new studies based on emerging data
relevant to the cogent argument. The Guidelines should not be inflexible prerequisites, he said.
Lauren Zeise: While linear defaults are conservative and protective, the current draft
Guidelines miss important exposures very early in life and very late in life when the difference in
risk can be as great as an order of magnitude. One major assumption implicit in the Guidelines
that may lead to miscalculation of risk is the homogeneity assumption, she said. The Guidelines
should incorporate a framework that allows adjustments for heterogeneities such as differences
across the population, polymorphisms, and other variability within species, timing of exposure,
and the impact of lifetime dose. The Guidelines'.assumption about lifetime average dose ignores
important information about timing of exposure, she said. Evaluation of epidemiological data
may reveal an environmental role in childhood cancers that now have no known cause. With
respect to the cogent biological rationale, it is critical to include data that make it possible to
calculate and integrate the effect of chronic background exposure, she said. The mode-of action
approach involves discussion of associations that support the hypothesis, she said. The
Guidelines should incorporate incentives, supported by Federal agencies, for a broader testing of
some hypotheses that are now employed in the mode-of-action approach.
Daniel Krewski: Dr. Krewski said the new Guidelines' emphasis on mode of action was
a significant improvement, but noted that developing sufficient information on specific modes of
action will be extremely difficult and emphasized that the Guidelines should encourage
development of methods to acquire this information. Assumptions about lifetime average daily
dose can lead to underestimations when early exposures are more important than later exposures
because of children's differential susceptibility. He noted that there is a useful body of literature
that describes tools which could be used to evaluate time-dependent exposures. He suggested that
these methods could be modified to incorporate susceptibilities as a function of.time. Inter-
individual variations and genetic susceptibilities may account for more than 10-fold differences
22
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
in risk, and are not be adequately accounted for in current animal bioassays, Dr. Krewski said,
and he suggested an additional 10-fold assumption about risks to children. He urged that risk
assessments be based on data related to in utero and perinatal exposures to assure that the critical
period of exposure is identified. Some carcinogens act through mechanisms that could invoke
either linear or nonlinear models for risk assessment, Dr. Krewski said, and the Guidelines
should incorporate models that allow for consideration of both the linear and nonlinear
contributions, not one or the other. Dr. Krewski noted that pharmacokinetic studies can be a
powerful tool to help identify specific susceptibilities in children. He also cautioned against ah
"across the board" confidence that an assumption of linearity offers the most conservative
evaluation of risk. Dr. Krewski said the Guidelines should also develop methods to take into
account the high risks that may be associated with human genetic factors alone or through their
interaction with varied environmental risk factors.
Frederica Perera: Molecular epidemiology studies in humans make it possible to
examine the issue of differential susceptibilities between the fetus and the young child and the
variability among young populations. These studies take advantage of biomarkers that can detect
molecular changes in samples of blood or other tissue. This approach allows a better
understanding of specific exposures, early damage, and susceptibility. Studies involving
polycyclic aromatic hydrocarbons and other aromatics such as pollution from coal burning,
traffic, and environmental tobacco smoke show that the fetus is at least 10 times more vulnerable
to damage than the mother. Other data from these studies show a differential susceptibility
among the children that is related to polymorphisms in the study population. Another study of
environmental tobacco smoke and preschool-age children also suggests that biomarkers can be
used to identify differential susceptibility related to ethnicity. Dr. Perera suggested that
biomarkers may provide a means to identify specific susceptibilities and to gather sufficient data
to develop probabilistic models that could lead to improved defaults.
23
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
Discussion
George Lucier noted that recent studies, including some involving biomarkers, have
found 100-fold variations in vulnerability between individuals. Michael Thun observed that the
presentations and discussions at the Workshop have focused on two juxtaposed issues:
On the one hand, a broad and "to the heart" issue of learning what causes, and what might
prevent, cancer in children.
On the other hand, the detailed mechanical considerations of risk assessment and
regulation.
In between, Dr. Thun noted, is a broad area of childhood cancer and its relationship with
infectious agents and pharmacological agents that may act more subtly than DBS. This area may
be beyond the province of the EPA, Dr. Thun said, and the Workshop should keep in mind that
the EPA is not going to be able to eliminate childhood cancers. Dr. Landrigan observed that from
70 to 80 percent of childhood cancers have unknown causes, but that it is clear that some are the
result of exposure to manmade synthetic chemicals. Human action is in part responsible for
causing these cancers, Dr. Landrigan said, and human actions such as the development of risk
assessment guidelines can be used to prevent them. Retha Newbold suggested that the Workshop
should not focus on childhood cancers to the exclusion of cancers that appear later in life but may
be the result of exposures that occurred in childhood.
Steven Galson asked Dr. Perera if sufficient data are available now to construct
probabilistic modeling that can be used to develop new defaults. Dr. Perera replied that in her
opinion enough data could now be gathered about the distribution of some genetic
polymorphisms and nutritional factors, as well as known factors related to gender and ethnicity,
to begin developing such models. Dr. Zeise added that some of these data could be integrated
24
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31, 2000
into a framework that helps describe individual differences and could be incorporated into risk
characterization.
Chris Portier said there is no convincing argument that the linear default is conservative
as a rule and that there is confusion in the Guidelines as to what the confidence bound derived
from animal estimates really represents in terms of protection of population-based risk compared
with variants around an estimate. He said the question of choosing a point of departure from the
observable response region to the extrapolation region is also not clearly answered in the
Guidelines. Dr. Portier said the concept of cogent biological rationale might be appropriate in
adults but cannot now be applied as a reason for moving away from the linear default in
assessing childhood cancer risks because not enough is known about mechanisms acting in
childhood cancers. He also urged the Agency to look at data that help elucidate the effect of long-
term versus short-term exposures and carcinogenesis in children.
Dr. Farland said the Guidelines should help provide the best possible judgment about risk
to the population, and these judgments must then be applied to decision-making. It is important
to prevent even an extremely small risk, he said, and the conservative nature of defaults makes
them an important tool in decision-making. He also said the Guidelines are open to the inclusion
of additional information and suggested that the Guidelines might incorporate language that
actively encourages the use of information such as data on biomarkers, epidemiologic
distributions, and ethnic factors. Dr. Portier noted that the draft Guidelines suggest that the
defaults would apply when there is neither adequate data nor a cogent biological rationale. He
urged that the Agency be very careful to support with data any action regarding children. He said
that he did not believe sufficient information was available to make a cogent biological argument
for the factors associating exposure to children's cancer.
25
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
TOPIC 2: ENHANCED USE OF TEST DATA RELATED TO CHILDREN'S CANCER
RISK
RochelleTyf
The currently-employed 2-generation reproductive toxicity test (OPPTS 870.3800)
monitors first-generation (F-l) animals exposed from the time they are gametes through
gestation, lactation, breeding, delivery, and weaning of second generation (F-2) animals. F-l and
F-2 generation animals are exposed "from womb to tomb" in this assay, Dr. Tyl said. She
described in detail the protocol and measured endpoints for evaluating effects on parental
animals and offspring. This study has the most potential for getting better information about
children's cancer risk than is now gathered. The study has the right exposure—spanning
development from gamete through adulthood—but gathers the wrong data for assessing
childhood cancer risks, Dr. Tyl said. She suggested that the study be extended to follow
development of F-2 animals beyond weaning of their offspring. This would allow detection of
long-latency cancers without the expense of carrying out separate chronic studies, she noted.
The prenatal developmental toxicity test (OPPTS 870.3700) involves exposure from
conception to birth. Because animals are necropsied at birth, this study captures only
developmental effects of the prenatal exposure and can not detect postnatal effects, Dr. Tyl
noted. Without major change in protocol, this test has very little value for assessing children's
cancer, she said.
The combined chronic/carcinogeniciry study (OPPTS 870.4300) involves exposure that
begins at age 6-8 weeks and continues through 18 months for mice or 24 months for rats. The 90-
day toxicity study (OPPTS 870.3100) involves exposure begun at 6-8 weeks and continued
through 13 weeks. Immunotoxicity studies (OPPTS 870.7800) begin exposure at 6-8 weeks and
continue through 28 days. These studies make it possible to detect impairment of cells involved
in immune response, and may be incorporated into the combined chronic/carcinogenicity and 90-
26
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31, 2000
toxicity studies. Metabolism and pharrnaeokinetics studies (OPPTS 870.7485) begin a 7-day
exposure at age 6-8 weeks and are conducted only on male animals. All of these studies begin
exposure on young adult animals and can therefore not contribute to assessing risks of exposure
during development or childhood, Dr. Tyl observed.
In conclusion Dr. Tyl said that the 2-generation studies, which involve exposure
beginning at implantation, hold the most promise for gathering information specific to childhood
risk. Following up F-2 generation animals through a chronic study would result in a study that
incorporates the appropriate exposure with long duration study. This would allow detection not
only of childhood cancers but also adult cancers initiated by childhood exposure.
Facilitated Discussion
Mark Miller: A systematic review of data gathered in other animal studies may reveal
timing and organ-specific information on mode of action that can be used to compare adults and
children. The single-exposure carcinogenesis data base may be one area that might be fruitfully
explored, Dr. Miller said. He suggested that reviewing existing research results to sort species by
chemical may help identify which species are best suited as models for specific chemicals. The
developing area of research into immune system effects should be integrated into testing for
cancer risk in children, he said. Precancerous conditions such as myeloplastic syndrome, which
progresses to acute myeloid leukemia, are frequently associated with specific genetic markers
and could reveal some associations between exposures and childhood cancer, Dr. Miller
suggested.
Paul Foster: Current reproductive and developmental studies involving prenatal and
juvenile exposure were not designed for cancer endpoints. With some modification these tests
might reveal early indicators of change related to cancer, but as they are now designed these tests
27
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
reveal the most relevant exposures but the least relevant endpoints for cancer risk, Dr. Foster
said. Among the current studies, data gathered in multigenerational studies have the most
potential for revealing childhood cancer risk, but the selective culling of animals in these studies
must be modified to include more, and more representative, animals per generation, he said.
Currently available "non-standard" data that may be collected include responses during specific
developmental windows of opportunity and hormone-like activity of possible carcinogens. Dr.
Foster also suggested that developmental stages of test animals compared with humans must be
considered. For example, he noted, early brain development that occurs prenatally in humans
takes place postnatally in rodents. Dr, Foster said that studies using transgenic rodents may
increase the sensitivity of the tests for specific cancers, but he cautioned that results obtained
from the study of increasingly sensitive rodents may have decreasing relevance to humans. There
is a huge opportunity to use the emerging knowledge of human and animal genomics to find
common mechanistic pathways for development of cancers, he said.
Dr. Foster said current prenatal developmental toxicity studies look at inappropriate
endpoints and are of no practical use in studying children's cancers, but that current
multigenerational studies could be modified to produce data relevant to childhood cancers.
Overcoming the limitations of current studies would require modifications such as determining
correct dose levels and increasing the number of animals studied from each litter, but extensive
modifications run the risk of making the studies too large and complex to be conducted
effectively, he said.
General Discussion
Dr. Goldman noted that the discussions of modifying current tests involve looking more
and more carefully at an increasingly homogeneous population of animals, and she contrasted
that with opinions voiced earlier in the Workshop suggesting that existing studies are too
28
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31, 2000
narrowly defined to capture the variability in exposure and susceptibility in children. One aspect
that needs to be more fully explored, she said, is the uniquely human characteristic of not breast-
feeding infants. Dr. Tyl noted that even the highly inbred rodent strains used in laboratory studies
do exhibit some variability, but agreed that it is small. It would not be possible, Dr. Tyl said, to
design animals studies that reflect the variabilities of the human population. George Lambert
agreed that the world of animal science does not reflect the conditions encountered by human
children. He suggested that this disparity argues for focusing on mechanism of action in
children's cancer studies. George Lucier suggested that the limited variability encountered in
animal studies could be examined more thoroughly to detect early markers that might be related
to variation. He also noted that a common mode of action, such as a receptor-mediated toxicant,
may produce different responses depending on the timing of exposure or the organ system
involved. Lucy Anderson said that studies that involve total life exposure involve influences
from conception through adulthood that may be additive, synergistic, or cancel out. Identifying
these effects would require different exposure patterns (preconception only, during gestation,
neonatal, and adult) to assure that critical effects are detected. She also agreed with earlier
comments that studies involving unculled litters studies could provide more information about
variability. Joseph DeGeorge emphasized the importance of timing exposures to coincide with
developmental stages, and noted that current rodent studies involve extrapolations from one
developmental stage to another. Extending these extrapolations still further, from rodents to
humans, must be done carefully, he noted. He suggested that more fully examining modes of
action can avoid some of the uncertainties of extrapolation.
William Farland noted that the regulatory toxicology tests are a very sensitive system that
is used to make judgments that rodent responses are relevant to humans for hazard identification
and that what is seen at high doses is relevant to low doses. He also said that animal studies are
beginning to identify biomarkers that can also be examined in human populations. He suggested
that information gathered about human biomarkers might be incorporated into animal models
29
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
through bioengineering. Dr. Landrigan noted that exposures in the toxicology testing mentioned
by Dr. Farland began when the animal had already reached adolescence and therefore is missing
important endpoints. He suggested that more meaningful data might result from tests in which
exposure began in utero and necropsy did not take place until the animal died naturally. Dr.
Farland responded that background effects of natural aging would complicate this type of
analysis, but agreed that tests in which exposure begins in adolescence miss the biological effects
of early exposure. David Wallinga questioned the usefulness of standard toxicity tests using
inbred animals to reflect the wide variability in humans. Moreover, he noted, the current
Guidelines implicitly assume that all humans are the same because there is no mention of
variabilities in susceptibility to carcinogens.
Les Robison suggested that the development of intermediate lesions might be a useful
precancerous marker for some childhood cancers. He also expressed concern about the reliability
of data derived from animal models to parallel the mechanisms and outcomes of human
childhood cancers.
Chris Portier noted that the mode of action approach in the Guidelines likely
accommodates most if not all of the concerns he had raised in earlier discussion about the
strength of available information in forming a cogent biological rationale. He supported the idea
of using the multigenerational study as a framework for developing a children's cancer bioassay,
but observed that the selection of some animals over others for study in each litter could result in
seriously overestimating or underestimating risk. The selection, he said, might be an unintended
result of culling, but also might be a result of the chemical itself. He also noted that studies
involving enough non-littermates to acquire meaningful data might require prohibitively large
numbers of animals. George Lambert noted that studies focusing on mechanism of action would
yield information relevant to risk for populations with different susceptibilities and inter-
individual variations.
30
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31, 2000
Chris Wilkinson noted that the discussion about problems posed by human heterogeneity
and the homogeneity of animal models overlooks the broader question of whether the rodent
model is adequate to assess risks in children in view of the fact that a newborn rat is essentially
equivalent to a human fetus. He suggested that developing new studies might be more productive
than modifying protocols of existing models because the huge numbers of animals needed for
study would pose a serious problem. He asked that the Workshop concentrate less on
environmental chemicals and consider ways to assess the risk of Pharmaceuticals, food additives,
or over-the-counter drugs, which may have substantial in utero impact. He said that
understanding modes or mechanisms of action makes it possible to plausibly extrapolate from
adult risk to children's risk.
David Goldsmith said the Workshop should be cautious in relying on developments that
may or may not derive from fuller knowledge of the human genome. He also suggested that
epidemiologic data can support dose-response data gathered in laboratory studies and should be
integrated more fully into the risk assessment process.
John Doe said that testing homogenous animal models at maximum dose can lead to false
positive associations, which is protective of public health. He also said it was important to
reiterate the point that the biggest difference between adult risk and children's risk is due to
exposure and not to hazard.
David Byrd said that there is a rich literature addressing the issue of the sensitivity of
false positive and false negative aspects of bioassays. Unfortunately, he said, the chemicals
represented in that data are not representative of the universe of environmental chemicals. He
also noted that variability represented in the animal species used in current bioassays is much
greater than the variability within the human population.
31
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
TOPIC 3: FUTURE DIRECTIONS FOR TOXICOLOGY TESTING TO ADDRESS
CHILDREN'S CANCER RISK
Retha Newbold
Dr. Newbold focused on research into the effects of DBS as an example of the cancer risk
associated with prenatal exposure to estrogenic chemicals. The developing organism is extremely
sensitive to estrogenic compounds, particularly during specific stages of development, and the
effects of exposure may not appear until much later in life. DBS was prescribed as safe and
effective to reduce risk of miscarriage, but now is known to have resulted in a low incidence of
vaginal cancer and a high incidence of male and female reproductive tract dysfunction on
offspring. Research into the effects of DBS demonstrated that a carcinogen can act across the
placenta, that its activity is different from other carcinogens, and that its effects in humans can be
accurately modeled in animals. Prenatal exposure of animal models results in developmental
effects in both male and female mice that closely parallel the effects found in humans, thereby
validating the experimental model as a means of predicting human disease. Neonatal studies, in
which exposure occurs during the first week of life, demonstrate that exposure to estrogenic
compounds during the period of uterine development—prenatal in humans and postnatal in
mice—is associated with uterine cancer. These studies confirm the critical role of timing of
exposure during developmental stages. To determine if the changes due to estrogenic exposure
could be transmitted to subsequent generation, researchers bred females exposed prenatally or
neonatally to control males and evaluated female F-2 offspring at maturity. Among the F-2
females, reproductive fertility was not effected, but the animals showed an increase in incidence
of reproductive tract tumors. Additional research is underway to determine the mechanisms
involved in these generational effects. Research advances developed through these studies of
estrogenic compounds may be applied to the development of more sensitive animal models of
other carcinogens.
32
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31, 2000
Facilitated Discussion
George Lambert: Risks to children from possible carcinogens can be studied in much
the same way as drugs are evaluated through pharmacokinetic and pharmacodynamic (PK/PD)
models, mechanisms of action, absorption, distribution, metabolism, and effect. These studies ,
can elucidate some of the differential susceptibilities between children and adults as well as those
due to interindividual variability. PK/PD studies are well suited to identifying differences
between the fetus and child or between child and adult. In humans, cells and tissues from
children and adults can be used to identify biomarkers that may be predictive. Dr. Lambert noted
that drugs, which are developed for use in a tightly defined population, are subjected to more
stringent examination than chemicals to which the whole population made be exposed through
the environment. He suggested that post-marketing surveillance of chemicals would reveal
patterns of distribution, exposure, accumulated body burdens, and adverse effects.
David Wallinga: Dr. Wallinga noted that the Workshop has been struggling to deal with
two different questions: what information is available and what information do we really need.
He said that the information available is limited and the data are poor. For example, he noted,
only a small percentage of the 80,000 registered industrial chemicals have been studied in even
limited detail for toxicity or carcinogenicity. He cited the National Research Council's 1993
observation that current testing protocols do not adequately address the toxicity and metabolism
of pesticides in neonate or adolescent animals. NRC also determined that infants and children are
more susceptible to risk than adults to the toxic effects of chemicals, even though chemical-
specific data may not be available. He noted the reasons for increased susceptibility and said that
the Guidelines should incorporate strong defaults assumptions and establish high hurdles to
abandoning those defaults. He said that the "cogent biological rationale" mentioned in the
Guidelines is not well defined and that the default assumptions in the Guidelines should be
closely examined to assure that they are sufficiently health-protective. For example, he said, the
33
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
Guidelines do not require data bearing on variability, interactions, or pharmacokinetic data in
developing animals. Dr. Wallinga recommended that future testing should:
Extend developmental toxicity tests beyond birth to account for latency.
Assess cancer risk from pre-conceptual exposure.
Look at effects of short-term carcinogen exposure during developmental windows.
• Require pharmacokinetics in immature animals.
Build developmental windows of vulnerability into the testing paradigm.
• Validate and test for endocrine disruption.
• Do semiquantitative assessments of cumulative risk.
Dr. Wallinga noted in closing that child-protective changes to the Guidelines can't wait,
as the Guidelines have been applied since 1996 to at least 45 pesticides and will, by the end of
FY 2001, have been used to establish new or revised IRIS assessments for at least 64 other
chemicals.
Joseph DeGeorge: Dr. DeGeorge said that it is important to consider the appropriateness
of the juvenile animal model before it is used to make predictions for children's risk assessment.
For example, he noted, if a toxicant needs activation by a metabolic process that the animal
model does not contain, the risk to children will be underestimated or missed. He said that if an
animal model is determined to be appropriate, the model must address exposure during the
proper organ developmental stage. The timing and method of exposure must be carefully selected
to assure that the effects of the chemical are isolated from confounding factors. For example, he
observed, exposure through lactation also involves exposure to metabolites. Dr. DeGeorge
recommended that more biomarkers be incorporated into the Guidelines, and that new
34
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31, 2000
biomarkers be added when they have been validated. He said that improving dose-response
assessment will be one of the most difficult challenges to the Guidelines because of the
complexity of exposure in humans. For example, he noted, the effective dose of the commonly
used nutritional supplement St. John's wort is reduce by 50 percent in persons taking protease
inhibitors, and grapefruit juice can increase the effective exposure to other Pharmaceuticals.
Discussion
Rochelle Tyl noted that much of the existing data on carcinogenicity have been based on
testing at the maximum tolerated dose. This means, as a practical matter, that the test animal's
metabolic capacity has been exceeded and no longer represents the effects in normal animals.
This may lead to quantitative differences in the measured outcomes and lead to inaccurate
conclusions about tested chemicals. John Doe said there is a practical problem associated with
test methods that rely on the induction of tumors as an endpoint. This approach would involve
huge experiments to assess the effect of in utero, early postnatal, postnatal through life, and
conventional dosing protocols. This complexity could be avoided through concentration on
identifying precursor events and other markers. The dilemma this poses, he noted, is that these
markers will not provide information that is as definitive as tumor initiation. Penelope Fenner-
Crisp noted that much of the revised and expanded testing being discussed would have to be
imposed on industries, a procedure that would require regulatory authority that does not now
exist. David Wallinga said that one of the purposes of defaults in risk assessment is to drive
research, and they should be used to drive research that generates data specific to children's risk.
George Lambert noted that the development of many FDA regulations has been driven by
Pharmaceuticals' risks to children, and suggested that the same concern might be brought to bear
on children's risks for cancer. Angelina Duggan noted that agricultural industries are beginning
to generate epidemiologic data that can be used to evaluate family exposures to chemicals that
are handled only by male farm workers. William Farland pointed out that the existing default
35
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
structure accounts for human variability through the implicit assumption that humans are at least
as sensitive as the most sensitive test animals.
TOPIC 4: EPIDEMIOLOGIC/MOLECULAR EPIDEMIOLOGY INFORMATION TO
ADDRESS CHILDREN'S CANCER RISK
Leslie Robison
Epidemiologic research is crucial to identifying risk and working to prevent childhood
cancer. These cancers are rare and unique, which makes epidemiologic study difficult. Acquiring
adequate etiologic data for childhood cancer will require a national effort to create a network for
research that would include:
A national registry of children with cancer for identifying environmental and other causes
of childhood cancer.
Building upon the unique national clinical trials system for treating children with cancer.
Identifying children at the time of diagnosis, allowing collection of tissue specimens.
Coordination of efforts with population-based cancer registries.
Support and facilitation of scientific studies of the highest merit by qualified investigators
to study causes of cancer in children.
The effort must be national in scope because of the differences between children's
cancers and those in adults and the relatively small numbers of children with any specific
diagnosis. Only a national effort would be able to compile enough data about the 8,700 cases of
childhood cancer diagnosed each year to make meaningful evaluations of specific cancer types. A
national network would make it possible to identify causes and to more fully understand known
risks factors, Dr. Robison said, and it would also lead to advances in molecular characterization
of tumors, exposure assessment methodology, and understanding genetic susceptibility. Among
36
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
the improved methodologies and technologies that might be developed through a national effort
are:
• GIS technologies for hypothesis generation and correlation.
• Sophisticated categorization of occupational categories relevant to specific exposures.
Exposure assessment through the ability to detect minute quantities of substances in
biological fluids and in the environment.
• Identification of biological markers of exposure and susceptibility.
Identification of potential genetic susceptibility factors.
A national effort would overcome the limitations of previous childhood cancer causation
studies and have the secondary benefit of making possible a study of patterns of care and
enhancing surveillance capabilities.
The national network initiative for children's cancers could use as its foundation existing
clinical trial cooperative groups such as the Children's Cancer Group and the Pediatric Oncology
Group, which represent more than 200 institutions throughout North America. These existing
groups, which will combine as the Children's Oncology group, have developed extensive
epidemiologic data on a variety of childhood cancer, but have not yet developed substantial
information on the etiologic of children's cancers. A framework for the structure and registration
protocol, as well as a projected development time line have been developed. When established,
the national network will make it possible to track the progression of pediatric cancer survivors
and examine the effects of medical exposures to therapeutic radiation, intermediate markers and
biomarkers of effect and progression, and cancer outcome.
37
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
Facilitated Discussion
Michael Thun: The cancer control community is interested in what epidemiologic
approaches will have the biggest effect in identifying the causes of childhood cancer and
preventing childhood cancer. The small relative numbers of childhood cancers severely limits
possibilities for epidemiologic study. For example, although there is a spike in the incidence of
acute lymphoblastic leukemia between ages 2 and 4, the average number of cases is fewer than
85 per million. A cohort of 1 million children enrolled at birth and followed until age 20 would
experience the following cancers:
• Leukemia
Acute Lymphoblastic 596
Acute Myeloid 154
• Lymphoma
Hodgkins 240
Non-Hodgkins 210
CNS - Astrocytoma 280
Thus, even a huge cohort study would not produce numbers large enough to provide
meaningful study of the incidence of the most common childhood cancers. An epidemiologic
study looking at genetic polymorphisms through relation of disease to a gene would require
between 2,000 and 3,000 cases to achieve enough statistical power to look at gene/environment
interaction.
Dr. Thun noted that understand the causes and means of preventing childhood cancers has
been a real, pervasive, and persistent concern in the public health community and among parents.
It is a problem that needs to be addressed across agencies, and should not be considered
separately within EPA or the National Cancer Institute, he observed. He recommended that
38
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
federal agencies jointly fund a data resource that could make headway against childhood cancers,
which, although rare, cause enormous grief.
Lucy Anderson: Animal models should be developed for studying factors such as
susceptibility and stage specificity, which are not typically covered in current animal bioassays,
Dr. Anderson said. In addition, historical literature should be investigated to gather data on stage
specificity, susceptibility factors, and other issues that are meaningful to childhood studies. New
studies to test putative associations, hypotheses, and the validity of biomarkers are needed. These
could be well-designed modifications of current studies or new protocols involving transgenics,
but they will only be carried out with government financial support, she said. The Guidelines
should incorporate studies that examine the role of fathers in children's risks, an issue has been
overlooked in the regulatory context even though epidemiological and animal evidence suggests
an important role of paternal exposure, she said. There is reason to believe that a qualitatively
novel mechanism exists to contribute to preconceptional carcinogenesis related to gene
expression, she said. These tests could be designed to detect the role of paternal exposures
through carcinogens in pesticides, drinking water contaminants, and tobacco smoke.
Peggy Reynolds: Evidence of an association between increased incidence of lung cancer
and early initiation of smoking may suggest evidence of mechanisms of early exposure as a cause
of later life cancers. A multicenter study of lung cancer in nonsmoking women, which investigated
the role of environmental tobacco smoke (ETS) in cancer among nonsmokers, found no increased
risk of childhood cancers associated with childhood ETS exposures, but found that for adult
cancers, women with childhood exposures had nearly twice the risk associations of women with
adult ETS exposures. This finding may illustrate the issue of "shelf life" as a factor in later life
development of cancers resulting from early life exposure, Dr. Reynolds noted.
39
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
Chris Portier: Establishing a cancer registry would be usefUl in understanding the
etiology of childhood cancer and, even if it is not environmentally related, will help us to
understand better the environmental issues associated with childhood cancer. Developing and
following large enough cohort studies of biomarkers will be very difficult. The difficulties are
more pronounced for cancers that occur in adults as the result of in utero, postnatal, or childhood
exposures.
General Discussion
Daniel Krewski commented on the value of nationwide registries to examine childhood
cancers, and discussed the types of information and biological specimens that are acquired in
Canada's cancer registry. Adult cancer registries might serve as models for childhood cancer
registries, which would not only be valuable in current studies but also represent an incalculable
resource for future research. Leslie Robison noted that the highest single age-specific rate for
childhood cancer is in the first year of life.
Some pediatric cancers have genetic origins, Dr. Robison said, but most are initiated in
utero. He added that the evidence for a paternal role in children's cancer is driven by data on
occupation, although some animal models show an association between preconceptional radiation
exposure and cancer in offspring. Dr. Robison said a national birth cohort study would not be
likely to make a meaningful contribution to the study of children's cancers but could reveal
patterns of association between childhood exposure and adult cancers. Moreover, he said, a birth
cohort would be invaluable as a means of tracking exposures.
William Farland and Philip Landrigan described the efforts of an interagency task force,
chaired by the Secretary of HHS, the Surgeon General, and the Administrator of the EPA, to
initiate a nationwide birth cohort that would register at least 100,000 ethnically and racially
40
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31, 2000
diverse children as early as possible in pregnancy. Statistical information and biological samples
would be obtained from the parents as well as the child, who would be followed with
standardized examination protocols to age 18. Many details remain to be worked out, Dr.
Landrigan said, but the effort has begun and has enormous promise as a means of increasing
understanding of the etiology of childhood diseases other than cancer. Dr. Robison noted that
although the birth cohort could not directly address childhood cancer as an outcome it would
provide data on exposure assessment that could have incalculable value to the ability to do cancer
related research in the future. Chris Portier said that a cancer registry would have more value to
childhood cancers than a birth cohort. Philip Landrigan repeated his earlier comment that it is
inarguable that children are more heavily exposed to carcinogens than are adults, that developing
organ systems are more vulnerable than fully formed systems, and that children have more years
of future life in which to develop cancers than do adults.
Joel Bender observed that the Workshop and similar discussions may not have been able
to reduce uncertainty about the Guidelines but have been successful in articulating a national
agenda to fill data gaps. A question that needs to be answered, he said, is whether the scientific
community is comfortable with the Guidelines until those gaps are filled. Lynn Goldman
expressed confidence that discussions would result in modifications to the Guidelines that
appropriately respond to concerns about pediatric cancer. David Goldsmith asked that serious
consideration be given, when establishing a birth cohort, to the disproportionate burden of severe
environmental and health problems borne by minority communities. He suggested that minority
children be oversampled as a method of assuring that their risks be fully incorporated into the
overall consideration of public health.
41
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
WORKSHOP SUMMARY
George Lucier
Dr. Lucier repeated the basic charge that the Workshop should serve as an opportunity to
help assure that the Guidelines for Cancer Risk Assessment accommodate as much information
as possible to adequately identify and assess risks to children. He noted the major themes that
emerged from the Workshop presentations and discussions:
Children should not be considered a special population; rather, childhood should be
considered a stage of development characterized by increased sensitivity to chemical
exposure.
• Evaluating margins of exposure requires better information on external exposures from
environmental sources such as air, food, and drinking water. Data on internal exposures,
as measured in blood or urine levels, is often not available for childhood exposures.
Pharmacokinetic and pharmacodynamic models can help elucidate childhood exposures.
All of this information should be considered with respect to exposure during critical
"windows" of organ development and cell differentiation.
Important sensitivity factors include developmental stage and genetic predisposition.
Much information that will emerge from refined test protocols, human genomics studies,
and other resources will bear on these issues, and organizing this new information into a
coherent picture of childhood cancer will be a challenging task. Other important
sensitivity issues relate to nutrition, "shelf life" (i.e., when exposure occurs during
childhood, there is a long latency period available for cancers to develop), and
interactions between multiple environmental, physiological, and genetic factors.
Mode of action (MOA) assessments should evaluate conditions in which different modes
lead to different risks (children compared with adults, for example) as well as when the
same mode leads to different risks. Genotoxic and nongenotoxic are oversimplified
distinctions when considering mode of action in the context of deciding whether to use
linear or nonlinear assumptions about risk. Background exposures must be taken into
account when considering MO A, as multiple factors with the same mode of action could
result in a cumulative risk.
42
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31, 2000
• Dose-response studies are difficult but important, and should take into account
background exposures, variability, dose selection and timing, and the applicability of
surrogate markers and early indicators.
Uncertainty factors include the realization that using defaults to account for variability is
more difficult than using them to account for species differences. The 10-fold safety
factors now used to account for species differences are public-health-protective, but may
not accurately reflect interindividual differences or differential genetic susceptibility;
therefore additional safety margins may be needed. Additional factors may be needed to
adequately assess children's exposure, and linear assumptions may not always be the
most conservative.
• Guidelines should not be overly prescriptive and should be able to stand the test of time.
Guidelines that include highly detailed descriptions of their application may prove to be
too inflexible to allow new approaches and new models to be used as they become
available.
• Regarding a cogent biological rationale that justifies an assumption other than the
guideline default: for children, the bar should be high enough to be public-health-
protective, and the models used will require rigorous peer review.
Discussion
Rochelle Tyl asked what could be expected as a result of the Workshop. Dr. Lucier
replied that the Workshop's purpose was to identify what information needs to be captured by the
Guidelines in order to more fully understand children's risks. This information, he said, would be
used to revise the current draft Guidelines. William Farland noted that the Workshop discussion
provided valuable insight into the need for data collection and generated ideas about revisions to
epidemiologic studies and laboratory testing protocols that may be incorporated into testing
Guidelines to assure that they fully address risks unique to children. He also noted that the
discussions of improved testing protocols would be valuable in improving other Agency
guidelines. David Wallinga urged that the Guidelines incorporate an explicit mention of
principles particular to dose-response or mode of action, such as a statement that, in the absence
of data to the contrary, mode of action for children is presumed to be different than for adults. A
43
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
similar principle could be framed for children's exposures, Dr. Wallinga said. Dr. Lucier
suggested that it would be necessary to develop credible models in order to obtain sufficient data
to fill the gaps in current understanding of both susceptibility and exposure.
Dr. Lambert noted that, for many children's cancers, susceptibility may be far more
significant than exposure. Michael Firestone urged that the effort to improve testing
methodologies be continued beyond the revision to the Guidelines and that the issues of windows
of susceptibility and differential exposure be built into the Guideline discussions of cogent
biological rationale. Rochelle Tyl suggested that the Guidelines incorporate a specific definition
of cogent biological rationale. Each researchers has their own perception of what the term means,
she said, but the Guidelines would benefit from a definition that articulated details such as types
of studies and timing of exposures that would satisfy the requirements of producing a cogent
rationale. Lynn Goldman responded that the Guidelines should be less specific, not more
detailed. Spelling out specifically how the Guidelines are to be applied today will make it more
difficult to apply them more effectively in the future, she said. Abraham Tobia replied that the
Guidelines do need some detailed and specific guidance to the research community, perhaps as
appendices or illustrative examples. Such examples, he said, would provide insight into the
thinking that underlies the Guidelines and therefore provide valuable direction to researchers,
particularly in regulated industry. Dr. Farland noted that the Workshop discussions had generated
a much clearer perception of what information is needed to clarify issues such as cogent
biological rationale, and suggested that the Guidelines might benefit from inclusion of a case
study or other example that more clearly articulates the concept of cogent biological rationale.
John Doe noted that new testing regimes need to be developed in response to the need for
specific types of new information, and said that these new testing protocols should replace older
study designs rather than being presented as add-ons to an already-burdened testing structure. Dr.
Tobia said that any new studies must be rigorously tested and validated before they are
44
-------
Summary of EPA Workshop on Information Needs to Address Children's Cancer Risk
Final Report, October 31,2000
incorporated into the Guidelines. He noted that the add-ons to multigenerational studies
discussed earlier in the Workshop could easily lead to a protocol that requires far higher numbers
of animals and round-the-clock attention by technicians and would make testing too complex and
prohibitively expensive. Dr. Goldman and Retha Newbold both cautioned that animal welfare
issues could become an important consideration if testing protocols become more complex, and
suggested that the Agency pursue development of studies that reduce the need for animal models.
Dr. Landrigan urged that children's risk should be an integral part of the overall cancer
risk Guidelines, not an afterthought. Dr. Farland said that he expected the final Guidelines to
include an explicit mention of children's risks. He noted that the purpose of the Guidelines was
protection of public health and that protection of children as among the most vulnerable is an
integral component of that goal. This point will be clearly made in the Guidelines, he said.
Dr. Landrigan and Dr. Goldman thanked the Workshop participants and observers for
their efforts and contributions. Dr. Farland expressed EPA's gratitude to the Workshop chair and
cosponsor and closed the Workshop by saying that all the participants could leave knowing that
they had contributed to progress.
45
-------
-------
APPENDIX A
WORKSHOP PARTICIPANTS
-------
-------
&EPA
United States
Environmental Protection Agency
Office of Research and Development
National Institute of
Environmental Health
Sciences
Workshop on Information
Needs to Address
Children's Cancer Risk
Holiday Inn Arlington at Ballston
Arlington, VA
March 30-31, 2000
Final List of Invited Participants
Lucy Anderson
Chief, Perinatal Carcinogenesis Section
Laboratory of Comparative Carcinogenesis
Frederick Cancer Research &
Development Center
National Cancer Institute
7th Street EXT.
Building 538 - Room 205-B
Frederick, MD 21702
301-846-5600
E-mail: andersol@ncifcrf.gov
Joseph DeGeorge
Center for Drug Evaluation & Research
Food and Drug Administration
Department of Health & Human Services
1451 Rockville Pike
HFD 024 WOC II - Room 6067
Rockville, MD 20852
301-594-5476
Fax: 301-594-5147
E-mail: degeorge@cder.fda.gov
William Farland
Director
National Center for Environmental Assessment
Office of Research & Development
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennslyvania Avenue, NW(8601 D)
Washington, DC 20460
202-564-3322
Fax: 202-565-0090
E-mail: farland.william@epa.gov
CLj) Printed on Recycled Paper
Penelope Fenner-Crisp
Office of Pesticide Programs
Office of Prevention, Pesticides &
Toxic Substances
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennslyvania Avenue, NW(7501 C)
Washington, DC 20460
703-605-0654
E-mail: fenner-crisp.penelope@epa.gov
Michael Firestone
Office of Children's Health Protection
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, NW(1107)
Washington, DC 20460
202-260-7778
Fax: 202-260-4103
E-mail: firestone.michael@epa.gov
Paul Foster
Chemical Industry Institute of Toxicology
6 Davis Drive
Research Triangle Park, NC 27709
919-558-1274
Fax: 919-558-1300
E-mail: foster@ciit.org
A-3
-------
Steven Galson
Office of Science Coordination & Policy
Office of Prevention, Pesticides &
Toxic Substances
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennslyvania Avenue, NW(7201)
Washington, DC 20460
202-260-6900
Fax: 202-401-0849
E-mail: galson.steve@epa.gov
Lynn Goldman
School of Public Health
Johns Hopkins University
624 North Broadway - Room 441
Baltimore, MD 21205
410-614-9301
Fax:410-659-2699
E-mail: lgoldman@jhsph.edu
Daniel Krewski
Professor
Department of Medicine and
Department of Epidemiology &
Community Medicine
University of Ottawa
451 Smyth Road - Room 3229C
Ottawa. ON K1H8M5
CANADA
613-562-5800, Ext.: 8261
Fax:613-562-5465
E-mail: dkrewski@uottawa.ca
George Lambert
Environmental & Occupational
Health Sciences Institute
681 Frelinghuysen Road - #406
Piscataway, NJ 08854
732-445-0174
Fax:732-445-0119
E-mail: glambert@umdnj.edu
Philip Landrigan
Department of Community &
Preventive Medicine
Mount Sinai School of Medicine
101st Street-#10(BSB)
Between Fifth & Madison Avenues
New York, NY 10029
212-241-4804
Fax: 212-996-0407
E-mail: phil.landrigan@niehs.nih.gov
George Lucier
National Institute of Environmental
Health Sciences
National Institutes of Health
111 T.W. Alexander Drive
Building 101 - Room A330
Research Triangle Park, NC 27709
919-541-3802
Fax: 919-541-3647
E-mail: lucier@niehs.nih.gov
Mark Miller
Office of Environmental Health
Hazard Assessment
California Environmental Protection Agency
1515 Clay Street - 16th Floor
Oakland, CA 94612
510-622-3159
Fax: 510-622-3210
E-mail: mmiller@oehha.ca.gov
Retha Newbold
National Institute of Environmental
Health Sciences
National Institutes of Health
111 T.W. Alexander Drive
Building 101 - Room 344
Research Triangle Park, NC 27709
919-541-0738
Fax:919-541-4634
E-mail: newbold1@niehs.nih.gov
Frederica Perera
Columbia University
40 East 94th Street - #30A
New York, NY 10128
212-304-7280
Fax: 212-544-1943
E-mail: fpp1@columbia.edu
Chris Portier
National Institute of Environmental
Health Sciences
National Institutes of Health
111 T.W. Alexander Drive
Building 101 - Room A344
Research Triangle Park, NC 27709
919-541-4999
Fax: 919-541-1479
E-mail: portier@niehs.nih.gov
A-4
-------
Leslie Robison
University of Minnesota
425 East River Road
Cancer Center Research Building - Room 554
Minneapolis, MN 55455
612-626-2902
Fax: 612-626-4842
E-mail: robison@epi.umn.edu
Michael Thun
Vice President, Epidemiology &
Surveillance Research
American Cancer Society
1599 Clifton Road, NE
Atlanta, GA 30329
404-329-5747
Fax: 404-327-6450
E-mail: mthun@cancer.org
Abraham Tobia
Manager, Toxicology
Toxicology Group
Aventis CropScience
2 T.W. Alexander Drive
Research Triangle Park, NC 27709
919-549-2213
Fax: 919-549-2925
E-mail: abe.tobia@aventis.com
Rochelle Tyl
Research Triangle Institute
3040 Cornwallis Road (HLB-245)
Rsearch Triangle Park, NC 27709
919-541-5972
Fax: 919-541-5956
E-mail: rwt@rti.org
David Wallinga
Senior Scientist
Natural Resources Defense Council
1200 New York Avenue, NW - Suite 400
Washington, DC 20005
202-289-2376
Fax: 202-289-1060
E-mail: dwallinga@nrdc.org
Lauren Zeise
Chief, Reproductive & Cancer
Hazard Assessment
Office of Environmental Health
Hazard Assessment
California Environmental Protection Agency
1515 Clay Street-16th floor
Oakland, CA 94612
510-622-319,5
Fax:510-622-3211
E-mail: lzeise@oehha.ca.gov
A-5
-------
-------
APPENDIX B
LIST OF OBSERVERS
-------
-------
SrEPA
United States
Environmental Protection Agency ,
Office of Research and Development
National Institute of
Environmental Health
Sciences
Workshop on Information
Needs to Address
Children's Cancer Risk
Holiday Inn Arlington at Ballston
Arlington, VA
March 30-31,2000
Final List of Observers
Daniel Axelrad
Analyst
Office of Policy
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennslyvania Avenue, NW(2174)
Washington, DC 20460
202-260-9363
Fax: 202-260-0512
E-mail: axelrad.daniel@epa.gov
Karl Baetcke
Senior Scientist
Office of Pesticides Programs
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, NW (7509C)
Washington, DC 20460
703-305-7397
Fax:703-605-0646
E-mail: baetcke.karl@epa.gov
Donald Barnes
Director, Science Advisory Board
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennslyvania Avenue, NW (1400 A)
Washington, DC 20460
202-564-4533
Fax: 202-501-0323
E-mail: barnes.don@epa.gov
Ted Barrera
President
Barrera Associates, Inc.
733 15th Street, NW- Suite 1120
Washington, DC 20005
202-638-6631
Fax: 202-638-4063
Lisa Barrera
Senior Vice President
Barrera Associates, Inc.
733 15th Street, NW- Suite 1120
Washington, DC 20005
202-638-6631
Fax: 202-638-4063
Joel Bender
4601 Ginger Trail
Toldeo, OH 43623
419-472-8390
Marilyn Brower
Office of Research & Development
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, NW(8601 D)
Washington, DC 20460
202-564-3363
Fax: 202-565-0062
E-mail: brower.marilyn@epa.gov
i Printed on Recycled Paper
B-3
-------
Daniel Byrd
President
CTRAPS
560 N Street, SW - Suite N-707
Washington, DC 20024
202-484-7707
Fax: 202-484-0616
E-mail: ctraps@radix.net
Gail Charnley
Health Risk Strategies
826 A Street, SE
Washington, DC 20003
202-543-2408
Fax:202-543-3019
E-mail: healthrisk@aol.com
David Chen
Office of Children's Health Protection •
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, NW (1107)
Washington, DC 20460
202-260-7677
E-mail: chen.david@epa.gov
David Clarke
Senior Director
Regulatory Reinvention and Legal Reform Team
Chemical Manufacturers Association
1300 Wilson Boulevard
Arlington, VA 22209
703-741-5160
Fax: 703-741-6092
E-mail: david_clarke@cmahq.com
Jim Cogliano
Branch Chief
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, NW(8623 D)
Washington, DC 20460
202-564-3269
Fax: 202-565-0079
E-mail: cogliano.jim@epa.gov.
Vicki Dellarco
Senior Scientist
Office of Pesticides Programs
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, NW(7509C)
Washington, DC 20460
703-305-1803
Fax: 703-605-0646
E-mail: dellarco.vicki@epa.gov
John Doe
Principle lexicologist
Zeneca CTL
Alderly Park, Macclesfield
Cheshire SK104TJ
UNITED KINGDOM
162-551-4556
Fax: 162-558-2715
E-mail: john.doe@ctl.zeneca.com
Paul Dugard
Halogenated Solvents Industry Alliance
2001 L Street, NW - Suite 506A
Washington, DC 20036
202-775-0232
Fax: 202-833-0381
E-mail: pdugard@hsia.org
Angelina Duggan
Director of Science Policy
American Crop Protection Association
1156 15th Street, NW - Suite 400
Washington, DC 20005
202-872-3885
Fax: 202-463-0474
E-mail: angelina@acpa.org
Alvin Edwards
Program Analyst
Office of Resource Management
Office of Research & Development
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, NW (8102 R)
Washington, DC 20460
202-564-6706
Fax: 202-565-2908
E-mail: edwards.alvin@epa.gov
B-4
-------
Elaine Francis
National Program Director for
Endocrine Disrupters Research
Office of Research & Development
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, NW(8701 R)
Washington, DC 20460
202-564-6789
Fax: 202-565-2444
E-mail: francis.elaine@epa.gov
Steve Gibb
Managing Editor
Risk Policy Report
1225 Jefferson Davis Highway - Suite 1400
Arlington, VA 22202
703-416-8578
Fax: 703-416-8543
E-mail: steve.gibb@iwpnews.com
David Goldsmith
Associate Research Professor
Department of Environmental &
Occupational Health
George Washington University Medical Center
2300 K Street, NW - Suite 201
Washington, DC 20037
202-994-1734
Fax:202-994-0011
E-mail: eohdfg@gwumc.edu
Utpal Gupta
Project Leader, Drug Safety Evaluation
Central Research Division
Pfizer, Inc.
Eastern Point Road
Groton, CT 06340
860-441-1864
Fax: 860-441-0438
E-mail: utpal_gupta@groton.pfizer.com
Karen Hammerstrom
Acting Division Director
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, NW (8601D)
Washington, DC 20460
202-564-3258
Fax: 202-565-0079
E-mail: hammerstrom.karen@epa.gov
Karen Hopfl-Harris
Associate Director for Policy
Physicians for Social Responsibility
1101 14th Street, NW- Suite 700
Washington, DC 20005
202-898-0150, Ext: 228
Fax: 202-898-0172
E-mail: khopfl@psr.org
Ann Hwang
Environment & Health Program Associate
Physicians for Social Responsibility
1101 14th Street, NW - Suite 700
Washington, DC 20005
202-898-0150, Ext: 223
Fax: 202-898-0172
E-mail: ahwang@psr.org
Sam Kacew
Department of Pharmacology
University of Ottawa
451 Smyth Road
Ottawa, Ontario K1H 8M5
CANADA
613-562-5800, Ext.: 8357
Carole Kimmel
Senior Scientist
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennslyvania Avenue, NW (8623 D)
Washington, DC 20460
202-564-3307
Fax: 202-565-0050
E-mail: kimmel.carole@epa.gov
Jane Koska
Analyst
Shook, Hardy & Bacon
600 14th Street, NW- Suite 800
Washington, DC 20005
202-783-8400
Fax:202-783-4211
B-5
-------
Arnold Kuzmack
Senior Science Advisor
Office of Water
Office of Science & Technology
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, NW(4301)
Washington, DC 20460
202-260-5821
Fax: 202-260-5394
E-mail: kuzmack.arnold@epa.gov
David Lai
Toxicologist
Office of Pollution, Prevention & Toxics
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, NW (7403)
Washington, DC 20460
202-260-6222
Fax: 202-260-1279
E-mail: !ai.david@epa.gov
Joan LaRock
President
Government Relations
LaRock Associates, Inc.'
6728 Baron Road
McLean, VA 22101
703-556-3324
Fax:703-734-7763
E-mail: joanlarock@erols.com
Elizabeth Margosches
Office of Pollution Prevention & Toxics
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennslyvania Avenue, NW(7403)
Washington, DC 20460
202-260-1511
Fax: 202-260-1279
E-mail: margosches.elizabeth@epa.gov
Carl Mazza
Science Advisor
Office of Air and Radiation
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, NW(6101A)
Washington, DC 20460
202-564-7427
Fax: 202-564-1407
E-mail: mazza.carl@epa.gov
Ray McAllister
Senior Director of Science & Regulatory Policy
American Crop Protection Association
1156 15th Street, NW - Suite 400
Washington, DC 20005
202-872-3874
Fax:202-463-0474
E-mail: ray@acpa.org
Robert McGaughy
Senior Scientist
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, NW (8623 D)
Washington, DC 20460
202-564-3244
Fax: 202-565-0078
E-mail: mcgaughy.robert@epa.gov
Rita Monroy
Project Director
National Alliance for Hispanic Health
1501 16th Street, NW
Washington, DC 20036 .
202-797-4334
Fax:202-797-4353
E-mail: rmonroy@hispanichealth.org
Ed Norman
Head
Children's Environmental Health Branch
Environmental Health Services Section
North Carolina Department of Environment
and Natural Resources
1632 Mail Service Center
Raleigh, NC 27699-1632
919-715-3293
Fax:919-715-4739
E-mail: ed.norman@ncmail.net
Pat Phibbs
Reporter
The Bureau of National Affairs, Inc.
1231 25th Street, NW
Washington, DC 20037
202-452-4106
Fax: 202-452-7891
E-mail: pphibbs@bna.com
B-6
-------
Peggy Reynolds
Chief, Environmental Epidemiology Section
Environmental Health Investigations Branch
Division of Environmental &
Occupational Disease Control
California Environmental Protection Agency
1515 Clay Street - Suite 1700
Oakland, CA 94612
570-622-4500
Fax: 510-622-4505
E-mail: preynold@dhs.ca.gov
Cindy Richard
Senior Scientist
Achieva
813 Frederick Road - Suite 200
Baltimore, MD 21228
410-788-1992
Fax:410-788-1993
E-mail: cindy@achievainc.com
Bruce Rodan
Senior Scientist
National Center for Environmental Assessment
Office of Research & Development
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avene, NW(8601 D)
Washington, DC 20460
202-564-3329
Fax: 202-565-0066
E-mail: rodan.bruce@epa.gov
Brianne Schaffert
Aidan O'Neil Foundation for the
Prevention of Childhood Cancer
1225 Cork Drive
Papillion, NE 68046
402-339-6411
Fax:402-399-1575
E-mail: saranrobo@home.com
Scott Schwenk
Office of Research & Development
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennslyvania Avenue, NW(8101R)
Washington, DC 20460
202-564-6667
Fax: 202-565-2431
E-mail: schwenk.scott@epa.gov
Bob Sonawane
Branch Chief
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, NW (8623 D)
Washington, DC 20460
202-564-3292
Fax: 202-565-0079
E-mail: sonawane.bob@epa.gov
Anne Travisano
Student Nurse
School of Nursing
Georgetown University
4547 Indian Rock Terrace, NW
Washington, DC 20007
202-333-3227
E-mail: travisaa@gusun.georgetown.edu
Linda C. Tuxen
Office of Research & Development
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, NW(8601 D)
Washington, DC 20460
202-564-3332
Fax: 202-565-0090
E-mail: tuxen.linda@epa.gov
Vanessa Vu
Associate Director for Health
Office of Research & Development
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, NW(8601 D)
Washington, DC 20460
202-564-3282
E-mail: vu.vanessa@epa.gov
Megan Wallace
Office of Air and Radiation
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avene, NW (6101 A)
Washington, DC 20460
202-564-7426
E-mail: wallace.megan@epa.gov
B-7
-------
Joanne Watters
Biologist
Novigen Sciences, Inc.
1730 Rhode Island Avenue, NW- Suite 1100
Washington, DC 20036
202-293-5374
Fax: 202-293-5377
E-mail: jwatters@novigensci.com
Chris Wilkinson
JSC, Inc.
1525 Wilson Boulevard - Suite 600
Arlington, VA 22209
703-312-8518
Fax: 703-527-5477
E-mail: chrisw@jscinc.com
Eric Wilson
Researcher
PETA
3435 R Street, NW-#8
Washington, DC 20007
202-337-1686
Fax:202-337-6097 .
E-mail: eric_w@peta-online.org
Jeanette Wiltse
Director
Office of Water
Health and Ecological Criteria Division
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennsylvania Avenue, NW (4304)
Washington, DC 20460
202-260-7317
Fax:202-260-1036
E-mail: wiltse.jeanette@epa.gov
William Wood
Executive Director, Risk Assessment Forum
Office of Research & Development
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Ariel Rios Building
1200 Pennslyvania Avenue, NW (8601 D)
Washington, DC 20460
202-564-3358
E-mail: wood.bill@epa.gov
Foster Young
Assistant Deputy Commissioner
Office of Policy & Intergovernmental Affairs
South Carolina Department of Health
and Environmental Control
P.O. Box 101106
Columbia, SC 29211-0106
803-898-0808
Fax: 803-898-0445
E-mail: youngfh@columb60.dhec.state.sc.us
Sheila Young
Student Nurse
School of Nursing
Georgetown University
1261 35th Street, NW
Washington, DC 20007
202-333-0351
E-mail: youngs2@gusun.georgetown.edu
B-8
-------
APPENDIX C
MEETING AGENDA
-------
-------
Workshop Agenda
Information Needs to Address Children's Cancer
Risk
Holiday Inn Arlington at Ballston
4610 North Fairfax Drive
Arlington, Virginia 22203
Thursday, March 30, 2000 and Friday, March 31, 2000
Thursday. March 30. 2000
6:00 - 7:00 PM
7:00-7:15 PM
7:15-7:30 PM
7:30 - 7:40 PM
7:40 - 7:50 PM
7:50-8:15PM
8:15-8:30 PM
8:30 - 8:55 PM
8:55-9:10 PM
9:10-9:20 PM
9:20 - 9:30 PM
Registration
Welcome by Sponsors
Introduction of Workshop Leads
Introductions
Charge to Workshop Participants
Discussion
Sensitivity of Children to
Environmental Toxicants
Discussion.
Exposure of Children to
Environmental Toxicants
Discussion
Comments from Observers
Session Wrap-up/Logistics
Friday. March 31. 2000
8:00 - 8:30 AM
8:30 - 8:55 AM
8:55 - 9:35 AM
9:35 - 9:45 AM
9:45-10:15 AM
&EPA
Registration
TOPIC 1: Current and Proposed
Approaches to Assessing
Children's Cancer Risk
Facilitated Discussion
Comments/Questions
Coffee Break
Eastern Research Group, Inc.
William Farland/George Lucier
Participants
Ramona Trovato, EPA
Participants
Lynn Goldman, John Hopkins
Participants
Phil Landrigan, Mount Sinai
Participants
Observers
George Lucier, NIEHS
Eastern Research Group, Inc.
William Farland, EPA
Abe Tobia, Aventis CropScience
Lauren Ziese, CalEPA
Dan Krewski, Univ. of Ottawa
Observers
C-3
-------
Friday. March 31 continued
10:15-10: 40 AM
10:40-11:50 AM
11:50-12:00 Noon
12:00-1:OOPM
1:00-1:25PM
1:25-2:05 PM
2:05-2:15 PM
2:15-2:45 PM
2:45-3:15PM
3:15-3:55 PM
3:55-4:15 PM
4:15-5:00 PM
TOPIC 2: Enhanced Use of Test
Data Related to Children's
Cancer Risk
Facilitated Discussion
Comments/Questions
Shelly Tyl, Research Triangle Inst.
Mark Miller, CalEPA
Paul Foster, CUT
Frederica Perera, Columbia Univ.
Observers
Lunch
TOPIC 3: Future Directions for Retha Newbold, NIEHS
Toxicology Testing to
Address Children's
Cancer Risk
Facilitated Discussion
Comments/Questions
Break
TOPIC 4: Epidemiologic/Molecular
Epidemiology Information
to Address Children's
Cancer Risk
Facilitated Discussion
Comments/Questions
Workshop Summary/Next Steps
George Lambert, EOHSI
David Wallinga, NRDC
Joseph DeGeorge, FDA
Observers
Les Robison, Univ. of Minnesota
(Ellen Silbergeld, Univ. of Maryland)
Michael Thun, Amer. Cancer Soc.
Lucy Anderson, NCI
Chris Poitier, NIEHS
Observers
George Lucier
C-4
-------
APPENDIX D
CHARGE TO PARTICIPANTS
-------
-------
«*• Charge to workshop Participants ««
Information Needs to Address Children's Cancer Risk
Thursday, March 30, 2000 and Friday, March 31, 2000
at the
Holiday Inn Arlington at Ballston
4610 North Fairfax Drive
Arlington, Virginia 22203
The purpose of the workshop is focused and derives from issues discussed in the EPA's 1999 Draft
Revised Guidelines for Carcinogen Risk Assessment.
What is the content of the ideal data set to adequately address children's cancer risk?
The workshop participants will focus on data needed for assessing the impact of childhood
(including in utero) exposures to carcinogens and the issues related to hazard identification and
dose-response analyses. The participants will address not only induction of childhood cancer,
but also increased risks of cancer during adulthood resulting from childhood exposure. As part
of this discussion, the participants also will be asked to consider how current protocols might be
redesigned to better answer questions related to children's cancer risk and what additional types
of data might be brought to bear on children's cancer risk assessment. This would include
information that is currently collected as well as data sets using new approaches.
What are the elements of a "cogent biological rationale," as presented in the draft revised cancer
guidelines (July 1999 Draft), for addressing modes-of-action for children's cancer?
Participants will address whether and how such a rationale can be made, which is sufficiently
health-protective of children, based on the kinds of data that are typically collected by and
available to Federal and state health science agencies at the present time. These might include
data on cancer mode-of-action, comparative pharmacokinetics and pharmacodynamics in adults
and children, rate and pattern of exposure in adults and children, etc. The background for these
discussions is the reality that chemical-specific data are often lacking to specifically address
children's cancer risk from environmental chemical exposures. As a consequence, the
assessment of children's risk is currently addressed by evaluations of traditional cancer
bioassays in mature animals using sensitive responders, comparative biochemistry and
physiology between adults and developing animals and humans, and public-health-protective
default positions in the absence of child-specific data.
It is expected that workshop discussions will be valuable to the general risk assessment community,
will provide input to Federal testing strategies for the future, and will inform the public dialogue
around children's health issues as they are addressed in the EPA's draft revised cancer guidelines.
A summary report of the perspectives and views coming out of this workshop will be published in
the peer-reviewed, scientific literature.
D-3
-------
-------
APPENDIX E
OVERHEADS USED IN THE PRESENTATIONS
-------
-------
Overheads from
Welcome and Charge to Workshop Participants
(William Farland and Michael Firestone)
E-3
-------
-------
Charge to the Children's Cancer
Workshop Participants
United Stales
Environmental Protection
toney
Of free of iChilclr
Ramona Trovato,
Director
Thursday, March 30,2000
EXECUTIVE ORDER
PROTECTION OF CHILDREN FROM ENVIRONMENTAL
HEALTH RISKS AND SAFETY RISKS
April 21, 1997
"... each Federal agency:
(a) shall make it a high priority to identify and
assess environmental health risks and safety risks
that may disproportionately affect children; and (b)
shall ensure that its policies, programs, activities,
and standards address disproportionate risks to
children that result from environmental health risks
or safety risks."
E-5
-------
Role of the Children's Cancer
Workshop Participants
The intent of EPA and NIEHS in sponsoring
this workshop is to obtain individual views
and perspectives from the participants.
Because we have not chartered this group
under the Federal Advisory Committee Act
(FACA), we are not seeking consensus
recommendation from the participants.
Focus of the Children's Cancer
Workshop
•The main focus of this workshop is to
discuss children's cancer risk
assessment and related data needs.
•A secondary focus is to address issues
that have arisen during review of EPA's
1999 Draft Revised Guidelines for
Carcinogen Risk Assessment.
E-6
-------
Specific Issues for Discussion
1. Characterizing the content of the ideal data
set to adequately address children's cancer
risk, focusing on data needed for assessing
the impact of childhood (including in utero)
exposures to carcinogens and the issues
related to hazard identification and dose-
response analyses
Specific Issues for Discussion
Addressing not only induction of childhood
cancer, but also increased risks of cancer
during adulthood resulting from childhood
exposure.
E-7
-------
Specific Issues for Discussion
Considering how current cancer bioassay
testing protocols might be redesigned to
better answer questions related to
children's cancer risk and what additional
types of data might be brought to bear on
children's cancer risk assessment.
Specific Issues for Discussion
Defining what are the elements of a
"cogent biological rationale," as presented
in the draft revised cancer guidelines, for
addressing modes-of-action for children's
cancer.
E-8
-------
Specific Issues for Discussion
5. Answering whether and how a "cogent"
rationale can be made, which is sufficiently
health-protective of children, based on the
kinds of data that are typically collected
by and available to Federal and state health
science agencies at the present time.
6.
Specific Issues for Discussion
Defining what additional data, such as
cancer mode-of-action, comparative
pharmacokinetics and pharmacodynamics
in adults and children, rate and pattern of
exposure in adults and children, etc., might
be useful in developing a "cogent"
rationale.
E-9
-------
Specific Issues for Discussion
7... Addressing whether the assessment of
children's risk, as it is currently conducted
by evaluations of traditional cancer
bioassays in mature animals using sensitive
responders, is sufficiently public-health-
protective in the absence of child-specific
data.
Communicating the Children's
Cancer Workshop Discourse
EPA and NIEHS will assist the workshop
participants in publishing a summary report
of the perspectives and views coming out of
this workshop in the peer-reviewed, scientific
literature.
E-10
-------
Overheads from
Sensitivity of Children to Environmental Toxicants
(Lynn Goldman)
E-ll
-------
-------
Sensitivity of Children to
Environmental Carcinogens
Lynn R. Goldman, M.D., M.P.H.
Public Health Issues
Childhood cancer and potential causes
Childhood carcinogenesis
E-13
-------
Childhood cancer issues
+ Low rate of occurrence and controversy
about trends
+ Unique cancer types
*• Very short latency
*• Many known genetic and familial
associations
+ High probability of gene/environment
etiologies
Childhood Cancer: Trends
+ Childhood cancer mortality is decreasing, yet:
+ Between 1973 and 1991 the rate of cancers for
children under the age of 15 increased at about
1% per year.*
+ Infants (< 12 months of age) are at relatively
high risk (22 cancers per 100,000 per year).
4 For infants, the increase between 1973 -1992
was 3% per year.
« The National Cancer Institute is expected to update these statistics in the very
near future.
E-14
-------
Childhood Cancer Trends
Year
-^-0-4 -«-"5-9" -*-'10-14" -B-15-19'
Childhood ALL and Brain cancer: 1973-96
Ra
E-15
-------
Childhood carcinogenesis
* Is childhood a time of greater susceptibility
to carcinogens?
+ Specific examples that raise concerns:
- DES exposures
- Radiation induced cancers (especially leukemia
and breast cancer)
- Smoking initiation and lifetime risk for lung
cancer
Environment and cancer
+ Probably a small percentage of cancer
solely or in part due to environment
(estimates range from 5-25%)
«• This is however a large public health
burden
4- 540,000 deaths from cancer 1998
4- Range of 27,000 to 134,000 environmental
E-16
-------
Susceptibility issues re: the
rodent bioassay model
* Genetic susceptibilities
- Inherited predisposition genes
- Polymorphisms in pharmokinetics affecting dose
* Reduced time to tumor (latency) resulting from
rapid growth of tissues
+ Persistence of effects (e.g. mutations and
alterations of cell differentiation)
+ Nutritional factors
* Hormonal factors
E-17
-------
-------
Overheads from
Children's Environmental Health
(Lynn Goldman)
E-19
-------
-------
Children's Environmental Health
Children are not little adults
A life cycle approach to understanding
environmental risks to kids
Birth defects as a possible early report
Other possible reports: childhood cancer
and childhood asthma
Children are not little adults
Exposure
Metabolism
Growth and Development
Shelf Life
E-21
-------
Children are more exposed
They breathe more air, drink more water
and eat more food, pound for pound, than
adults.
Children play close to the ground, where
some pollutants concentrate.
Normal hand-to-mouth activity exposes
kids to pollutants in dust and soil.
Children have different
metabolism
They may be more or less capable of
breaking down, excreting, inactivating, or
activating toxic substances. These
differences should be taken into account.
E-22
-------
Children's rapid growth and
development make them
vulnerable
Organ systems - central nervous system,
reproductive organs, immune system, and
lungs - are more susceptible to toxic
insult at various phases of development.
Children have a longer shelf life
• With more future years for disease to
express itself, diseases with long latency
periods or requiring chronic exposure may
have more serious impacts.
E-23
-------
Introduction - Dioxins & PCBs
PCB's = two linked phenyl rings + Cl
PCB's: industrial uses '30s - '70s
Dioxins & furans = contaminants from combustion &
manufacturing practices;
Long half lives (persistent), bioaccumulate, varying
toxicity
Most toxic is 2,3,7,8-TCDD (contaminant of Agent
Orange)
Example: Exposures to dioxins
from birth to age 25
Cumulative consumption of dioxins
newborn-25 years - males
10 15
Age (years)
20
25
Source: Patandin et a). Environmental Health
Perspectives, 107:1,1999.
E-24
-------
Example: Babies have higher
dioxin doses than adults
Dosage of dioxins (TEQ) by age- males
120
100
80
60
8 40
D
20
0
10 IS
Ago
20
Source: Potandin et a!. Environmental Health
Perspectives. 107:1,1999.
Intake Rates: Adults vs Children
• Drinking water
• Food
• Soil
• Air
E-25
-------
Drinking Water
Drinking water consumption by age (high end)
90 T—
age 1-10 age 11-19 adult
Age group
Food
200-
180-
160-
140-
a-120-
2 ioo
CT
E 80
60
20
0
Wn m-n B-
J J
i i , II • i i ,
<1 1-2 3-5 6-11 12-19 20-39 40-69 70 +
[•meat • dairy a eggs atlsh|
E-26
-------
Child (10kg)
Adult (70kg)
113 Mean B High end]
Air
Newborn Infant (1 yr) Children(1- Females Males (13-
12) (13-65) 65)
I Meter3/kg-day B Liters/kg-min |
E-27
-------
-------
Overheads from
Exposure of Children to Environmental Toxicants
(Philip Landrigan)
E-29
-------
-------
DIETS OF
INFANTS
AND
CHILDREN
1.2%Vover that'^ioa,
17.3% (the latter increase was ''
-
scended testes, inguinal hernia, and prenatal factors have been implicated
as possible risk factors, the cause of the trend is unknown. ' •'
Incidence rate per 100,000*
'6 •
White mule*
1 ' 1Mr~i iii »""-i — (—'i ..... •) — r— T-— i—-i — i — i — i — i — i "— i--r
1973 1975 1977 1979 1981 1983 1985 I9B7 1989 1991 1993 1995
Year of diagnosis
l Ca*crr Katti anJ Ruks. JAeditio*
E-31
-------
HTf POSFRQISS/EPISPSQ1 flS
TRENDS IN REPORTED INCIDENCE. BT QUHRTER OF BIRTH.
BIRTH DEFECTS MONITORING PROGRFW / CPHH
JHH 1970-DEC 1993
IBRTES PER 10.000 70TRL BIB1H31
„ -• . QUBRTERLT RHTES
a-TERB MOVING HVERflGE
70 71 72 73" 7« 75 78 !7T 7B 73 00 81 02 OS BM BS 88 87 SB 63 BO 31 BZ 93
! ' YEBB OF BlfltH
Incidence of Childhood Cancer
New Cases
1 per
100,000/yr
i 1972 1976 1980 1984 19813
Incidents of cancer In white children by calendar year of diagnosis, 1973-1988, United States, SEER
(data from Table 11-21 In the original report). The correlation coefficient, r, Is 0.542. j
E-32
-------
Prevalence Rate^ for Asthma by Age
and Year, United States, 1984-1994
Rate per
1000
population
'82 '83 '84 '85 '86 '87 '88 '89 '90 '91 '92 '93 '94 '95
Source: Centers for Disease Control & Prevention
Examples'of the " '•
New Pediatrio Morbidity
-------
The "New Pediatrics Morbidity!'
A range of chronic conditions of ;
complex and poorly defined origin
that account for most hpspitalizations
and deaths among American children
today. ;|
Chemicals with medium/high
potential human exposure:
proportion with minimum
screening data
E-34
-------
E-35
-------
% CoKguEsr cf PssmERCE ia. NEW YORK CITY ~
^crftShomlg the Deatft Rate i ' ' ' "'
REGOSTOEMDATIQMS Qi
THE HAS COW1M1TTEE i:
i1
Improvfe toxicity testing
Establish new, additional 10-fold j
uncertainty factor as a default
Collect better data on food consumption
^ ", - • "
Collect better data on pesticide residues
E-36
-------
:• • ' MAS COMMITTI
The Committee recommended that
changes be made in current regulatory
practice.
Tolerances must be based principally on
health considerations.
i
Estimates of exposure should reflect the
unique characteristics of infants and
children and should account for all dietary
and nondietary intake of pesticides. i
Children's exposures to pesticides are greater
pound-for-pound than those of adults.
Children are less well able than adults to detoxify
most pesticides.
Children's developing organ systems are highly
vulnerable to pesticides.
Children have more years qf future life in which to
develop chronic disease triggered by early
exposure. r^n
E-37
-------
EXAMPLES OF THE VULNERABILITY OF INFANTS AND CHILDREN TO
ENVIRONMENTAL TOXICANTS
• Increased risk of cancer following mitrosamine exposure (Peto)
• Increased risk of cancer following vinyl chloride exposure ( Drew)
• Decreased ability to detoxify organophosphates (Moser)
• Increased susceptibility to lead (Needleman)
• Increased susceptibility to alcohol (fetal alcohol syndrome)
• Thalidomide and phocomelia
• DBS and adenocarcinoma of vagina
. New possible: Neurodevelopmental toxicity of chlorpyifos (Whitney; Song;
Campbell)
E-38
-------
Congressional Charge
to the
NAS Committee on Pesticides & Children:
• Are children more heavily exposed than adults?
« Are children more susceptible to toxicity than adults?
• Do current laws and decision practices protect children?
Environmental Health
Threats To Children
E-39
-------
Executive Order on Protection of Children
From Environmental Health Risks and Safety
Risk
luuctl April 21.199T
THE WHITE HOUSE >
OHicrofiac Frai Secretary
Far Itnmciluic Rtteaj*
By Jh« authority vested in ow as President by ifae Ginitiliuicn and the U« 1 of ihe L'nittd
StattJ of America, u |i hereby ordered as faltowi:
JfCffan /. Policy
1.101 A growing body of Kicntitic knowtcdjc deewtuuates thai children nuy suffer
aiiKtxMrtioriUelv from environmental health nils and safety rijLi, The« mla ante
beJauirehildrca'i nwrolosfca!. urumnwlosical. dlfiudve. a»! other bodily tpwaa «
j^devrfocbic rftUOrcu «! more food, dri«k mor. lluidj. and bteail* mcr= ju in
praponioni.} UtcirboOy uvi^ht than adultr. chiIOren*« lira and vx i
(a) jhilt males it ahif h priority W identify and awo* cnvirocuwstal health rf*U and
riis thac raay tUtpraponiwaiely atTeu: chitdrcn: and
00 jhall cuure -Jut tU eolEciet. prograaH. arfvities. anJ lUcdanli address
n&s w children thai mult from «nviK>n. with its pnsvmoia.
Sff. 2. Definition*. The following Jcfintticu dull appl> » ih» order.
MOI. -Federal ak-«cy- nwans any authority of the Unit*! Sutw jhat is M asettcy under «
U.S.C, 35«ll) other than tfao« eocwdeicd to be. lodepenJeni nrsuUwty iS««»« "w^J4
Ifli XT: e»vere-d under iheauipicei of tbebepanaiett of Defense.
2-:- action* mean* any iubnantis-e action in a rulerr-afcing. teu&taJ
aflcr the date of th« cider ot for which a Notice of rrojuwea Rale.-nakins ^ pubtwhsd I jew
after the tlite of thli on!cr.ihat is likely to mull ma rule ihK cuy:
(a) br-«ocomtcal!* rigniGcaaf voter Exeswivc Order !2K6 (a ru!enu!di« that ha* as
annual elYe« on ihe'wonMny of SI 00 rrJIlion ar nwe or »ould adversely afScet in a
m.-ilRul «a* ihe ecenotnv. a *ecur of ihe «c9o«ay. producaviti'. compeUfion.]ow,
ertinirjaeai. public healih « safety, ocState, \cxai. or tribal so«t=ae!iO orwmrcu
Food Quality Protection Act9'1996 ;'
Highlights of Federal Food, Drug, and Cosmetic Act
a Health-baspd safety standard for pesticide residues in food
- Reasbnable certainty of no harm
n Special provisions for infants and children
n Limits consideration of benefits of pesticides
j i : H
n Tolerances reviewed within 10 years ;j
j :\
n Penalties for violations s;
a Provisions! for endocrine testing
E-40
-------
Overheads from
Topic 1: Current and Proposed Approaches to Assessing Children's Cancer Risk
(William Farland)
E-41
-------
-------
An EPA/NIEHS Workshop:
Information Needs to Address Children's Cancer Risk
March 30 and 31, 2000
An EPA/NIEHS Workshop:
Information Needs to Address
Children's Cancer Risk
March 30 and 31, 2000
TOPIC 1
Current and Proposed Approaches
to Assessing
Children's Cancer Risk
William H. Farland, Ph.D., Director
National Center for Environmental Assessment
Office of Research and Development
U:S. ENVIRONMENTAL PROTECTION AGENCY
Research +=* Assessment 3=± Management
EPA Scientific Research/
Data Collection
Risk Assessment
Animal Toxicolo
Clinical Studies
Epidemiology
Cell/Tissue
Experiments
Computational
Methods
Monitoring/
Surveillance
Dose-Response
Characterization
Hazard
cterization
Exposure
Characterizati
• Other Federal Agencies
• States/Local Agencies
•Acaderrria-
• industry
Public Interest/Environmental Groups
Non-ris
Analyses
External Input
into Research/
Assessment
William H. Farland, Ph.D., Director
Mailing Address:
National Center for Environmental Assessment (8601D)
Office of Research and Development
U.S. Environmental Protection Agency
1200 Pennsylvania Avenue, NW Ariel Rios Building
Washington, DC 20460
.E-43
Office Location:
808 17th street, NW
Suite 400
Washington, DC 20074
Tel: 202-564-3322
FAX: 202-565-0090
E-mail: farland.william@epa.gov
-------
An EPA/NIEHS Workshop:
Information Needs to Address Children's Cancer Risk
March 30 and 31, 2000
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
a
a
a
a
a
D
D
a
a
a
a
n
Recent Emphasis Focuses on
the Use of Mode of Action Data
"The quality of risk analysis will improve as the
quality of input improves. As we learn more
about biology, chemistry, physics, and
demography, we can make progressively better
assessments of the risks involved. Risk
assessment evolves continually, with
reevaluation as new models and data become
available."
"Science and Judgment in Risk
Assessment" (National Research
Council, 1994)
D
D
D
a
a
a
a
a
n
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
n
a
a
a
D
What Are Agency Risk
Assessment Guidelines?
^Statements of Agency policy regarding
principles, general approaches,
preferences, and default assumptions
that will be applied in Agency risk
assessments
S Not a Cookbook
SNota Regulation
William H. Farland, Ph.D., Director
Milling Addnss:
NaUonal Center (or Environmental Assessment (8601 D)
OIIIco of Research and Development
U.S. Environmental Protection Agency
1200 Pennsylvania Avenue, NW Ariel Rlos Building
Washington, DC 20460
-E-44-
Office Location:
808 17th street, NW
Suite 400
Washington, DC 20074
Tel: 202-564-3322
FAX: 202-565-0090
6-mail: farfand.william@epa.gov
-------
An EPA/NIEHS Workshop:
Information Needs to Address Children's Cancer Risk
March 30 and 31, 2000
D
a
a
a
a
a
n
a
a
a •
a
a
a
a
a
a
a
a
a
a
n
a
a
n
a
a.
a
a
a
Guidelines Are the
Product of:
> Several Agency Colloquia
> 3 External Peer Consultation
Workshops
> Peer Review Workshop
> Interagency Review (NIEHS,
OSHA,FDA, NSF, DOE, NIH)
> 3 Science Advisory Board Reviews
> Public comment (> 100 commenters)
D
n
n
D
n
a
n
n
a
a
a
a
a
a
n
a
a
a
a
a
a
n
a
a
a
a
a
D
a
Guidelines History
In September 1986, EPA published
Guidelines for Carcinogen Risk
Assessment (5 1 FR 33992)
1 986 Guidelines were based on the
principles and "state of the science" as
practiced in the 1970s and early 1980s
1 986 Guidelines allowed for flexibility
in application but little in the way of
guidance on when/how to depart from
default assumptions
William H. Farland, Ph.D., Director
Mailing Address:
National Canter for Environmental Assessment (8601 D)
Office of Research and Development
U.S. Environmental Protection Agency
1200 Pennsylvania Avenue, NW Ariel Rios Building
Washington, DC 20460
E-45.
Office Location:
808 17th street, NW
Suite 400
Washington, DC 20074
Tel: 202-564-3322
FAX: 202-565-0090
E-mail: farland.william@epa.gov
-------
An EPA/NIEHS Workshop:
Information Needs to Address Children's Cancer Risk
March 30 and 31, 2000
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
Q
a
D
D
a
a
D
Revision Directions for Risk
Assessment Guidelines
o Emphasize full characterization
o Expand role of mode of action
information (and, therefore,
biomarkers!)
o Use all information to design
dose response approach
o Two step dose response assessment
EPA's Guidelines for Carcinogen Risk
Assessment (Draft, July 1999) state:
"The interaction of the biology of the organism
and the chemical properties of the agent
determine whether there is an adverse effect.
Thus, mode-of-actjon analysis is based on
physical, chemical, and biological information
that helps to explain key events in an agent's
influence on development of tumors. The
entire range of information developed in the
assessment is reviewed to arrive at a reasoned
judgment."
William H. Fariand, Ph.D., Director
Mailing Address:
National Center for Environmental Assessment (8601 D)
Office of Research and Development
U.S. Environmental Protection Agency
1200 Pennsylvania Avenue, NW Ariel Rios Building
Washington, DC 20460
-E-46-
Office Location:
808 17th street, NW
Suite 400
Washington, DG 20074
Tel: 202-564-3322
FAX: 202-565-0090
E-mail: fariand.william@epa.gov
-------
An EPA/NIEHS Workshop:
Information Needs to Address Children's Cancer Risk
March 30 and 31, 2000
D
a
a
a
a
a
a
n
n
a -
a
a
a
a
a
n
a
a
a
a
a
a
a
a
a
a
a
n
n
Evolution of Hazard
Characterization
Hazard Identification j
through Traditional
Toxicologic Testing
Hazard Characterization
through Evaluation of
Mechanism(s) and
Biologically-Based Models
n
a
n
a
a
a
a
n
a
a
D
a
a
n
a
a
a
a
n
a
a
a
n
a
a
n
a
a
a
Mode of Action --
How does the chemical
produce Its effect?
Are there mechanistic data
to support this hypothesis?
Have other mechanistic
hypotheses been considered
and rejected?
William H. Fariand, Ph.D., Director
Mailing Address:
National Center for Environmental Assessment (8601 D)
Office of Research and Development
U.S. Environmental Protection Agency
1200 Pennsylvania Avenue, NW Ariel Rios Building
Washington, DC 20460
E-47
Office Location:
808 17th street, NW
Suite 400
Washington, DC 20074
Tel: 202-564-3322
FAX: 202-565-0090
E-mail: fartand.william@epa.gov
-------
An EPA/NIEHS Workshop:
Information Needs to Address Children's Cancer Risk
March 30 and 31, 2000
D
D
D
a
a
a
a
D
D
D
D
D
D
a
D
D
D
a
a
a
a
n
a
a
a
a
a
a
D
Use of Mode of Action Data
in Dose Response Assessment
o Construct a biologically-based or case
specific model
o Link dose response curve for precursor
effect to dose response for tumor effect
o Use dose response for other effect in
lieu of that for tumor effect if it is judged
to be a better measure of potential risk
o Use to inform assessment of possible
dose response in range of extrapolation
Dose Response Assessment
&ooafa.t^1a*&F&^^&&$:3rteL&.*
Range of
Extrapolallor
William H. Fartand, Ph.D.. Director
MsiHng Address:
National Center for Environmental Assessment (8601 D)
Office of Research and Development
U.S. Environmental Protection Agency
1200 Pennsylvania Avenue, NW Ariel Rios Building
Washington, DC 20460
Office Location:
808 17th street, NW
Suite 400
Washington, DC 20074
Tel: 202-564-3322
FAX: 202-565-0090
E-mail: fariand.william@epa.gov
-------
An EPA/NIEHS Workshop:
Information Needs to Address Children's Cancer Risk
March 30 and 31, 2000
D
D
n
n
D
n
n
n
n
n •
n
n
a
a
D
a
D
a
a
a
D
a
a
D
n
a
a
a
a
Decision Logic for
Dose Response Assessment
• Linear
+ DNA reactive or other evidence supporting linearity
+ not DNA reactive but insufficient data to characterize
a non-linear mode of action
• Non-linear
+ not DNA reactive or otherwise linear, and sufficient
data to characterize a non-linear mode of action
• Both
+ differing activity at different sites
+ complex activity needing both approaches
to describe
n
a
a
n
a
a
n
a
a
a
a
a
a
a
a
a
a
n
a
n
a
a
a
a
a
n
a
a
a
Linear Approach is Public
Health Conservative
• When the linear approach is used to
extrapolate several orders of magnitude
from observed data, EPA does not find
a need to add an adjustment for human
variability
• The NRC did not consider the overall
effect of extrapolation to low doses
+ PBPK models or scaling adjustment are
to be applied to account for interspecies
differences in toxicokinetics
William H. Fariand, Ph.D., Director
Mailing Address:
National Center for Environmental Assessment (8601 D)
Office of Research and Development
U.S. Environmental Protection Agency
1200 Pennsylvania Avenue, NW Ariel Rios Building
Washington, DC 20460
E-49
Office Location:
808 17th street, NW
Suite 400
Washington, DC 20074
Tel: 202-564-3322
FAX: 202-565-0090
E-mail: farland.william@epa.gov
-------
An EPA/NIEHS Workshop:
Information Needs to Address Children's Cancer Risk
March 30 and 31,2000
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
D
a
a
a
a
a
a
a
a
a
Linear Approach is Public Health
Conservative (continued)
* Low dose extrapolation is conducted on the
Point of Departure, the lower 95% confidence
limit on the lowest dose associated with
increased tumor response in the observation
range (e.g., LED10)
*• Under standard experimental conditions at
maximum tolerated dose, the test rodents are
considered to be stressed, not "average",
animals
*• The straight line approach results in risk
estimates similar to those by the IMS
procedure -- likely to overestimate at low
a
a
a
a
a
a
a
a
a
a
a
a
n
a
D
a
a
a
a
a
a
a
a
a
a
D
a
a
a
Human Variability —
Linear Approach
• The linear low dose extrapolation procedure
is assumed to be very public health
conservative without need for additional
factors to account for human variability
• The NRC (1994) suggested that "average
rodent" to "average human" extrapolation
may not account for variability adequately
when making estimates of individual risk
(maximum exposed individual(s) for a site
assessment), but is adequate for incidence
estimates for a population
William H. Farlaml. Ph.D.. Director
Mailing Address:
National Canter for Environmental Assessment (8601 D)
Office of Research and Development
U.S. Environmental Protection Agency
1200 Pennsylvania Avenue, NW Ariel RIos Building
Washington, DC 20460
E-50
Office Location:
808 17th street, NW
Suite 400
Washington, DC 20074
Tel: 202-564-3322
FAX: 202-565-0090
E-mail: farland.william@epa.gov
-------
An EPA/NIEHS Workshop:
Information Needs to Address Children's Cancer Risk
March 30 and 31, 2000
n
n
n
n
a
a
a
n
a
n
a
a
a
n
a
a
a
a
a
a
a
a
n
a
n
a
a
a
a
Human Variability—
MOE Analysis
IA margin of exposure (MOE) analysis is used
as a default approach when application of the
framework for mode of action (MOA) supports
a nonlinear MOA for adults and children
i Unless there is agent-specific data indicating
differential responses in children, the usual
factor for human variability is used and an
additional factor to protect children is not
applied (dose adjustment is made for
children)
n
a
n
a
a
n
a
n,
a
a
a
n
a
a
a
a
a-
a
n
a
a
n
a
n
a
a •
n
a
a
Proposed Approach to
Children's Risk
• Potential differences in exposure, dose,
and response between children and
adults
• A case-by-case approach based on
weight of evidence
• Default science policy positions and
procedures to be used in the absence
of data
William H. Farland, Ph.D., Director
Mailing Address:
Natfonal Center for Environmental Assessment (8601D)
Office of Research and Development
U.S. Environmental Protection Agency
1200 Pennsylvania Avenue, NW Ariel Rios Building
Washington, DC 20460
-E-51-
Office Location:
808 17th street, NW
Suite 400
Washington, DC 20074
Tel: 202-564-3322
FAX: 202-565-0090
E-mail: fariand.william@epa.gov
-------
An EPA/NIEHS Workshop:
Information Needs to Address Children's Cancer Risk
March 30 and 31, 2000
D
D
D
D
D
O
D
a
a
a
D
a
a
a
a
a
a
a
a
n
a
a
a
a
n
a
a
a
D
Early Life Sensitivity
• Guidelines call for a separate evaluation
when there are data indicating increased
sensitivity to early-life exposure (section
2.2.2.2)
• Case example (Appendix F) illustrates how
such information can be used to
characterize and estimate risks for early-
life exposure
• The proposed approach calculates an
adult and a childhood-specific unit risk
estimates
n
n
n
n
a
a
a
a
D
n
a
a
n
a
a
a
a
a
a
a
a
a
a
a
n
n
a
a
a
Childhood Risks -
Default Procedures
* Slope factors and unit risks for lifetime
exposure incorporate exposure factors
based on adults
• In the absence of childhood-specific unit
risk, and there is a need to assess risks
from childhood exposure, adult unit risk
is used after adjustment for potential
differences in dose between adults and
children (section 3.5.2)
W»!am H. Fartand. Ph.D.. Director
Mailing Address:
National Canter (or Environmental Assessment (8601 D)
Olllco of Research and Development
U.S. Environmental Protection Agency
1200 Pennsylvania Avenue, NW Ariel Rios Building
Washington. DC 20460
E-52-
Office Location:
808 17th street, NW
Suite 400
Washington, DC 20074
Tel: 202-564-3322
FAX: 202-565-0090
E-mail: fariand.william@epa.gov
-------
An EPA/NIEHS Workshop:
Information Needs to Address Children's Cancer Risk
March 30 and 31, 2000
Adult to Child Dose
Adjustments
• Default procedures for adult to child risk
adjustments based on differences in dose
•=> Oral slope factor - no adjustment is proposed
<=> Inhalation unit risk (gases) - adjustment based
on body weight and breathing rate
^ Drinking water unit risk - adjustment based on body
weight and drinking -water rate
m Guidance for inhaled particles, and dermal
exposure to be developed
• Are these default procedures appropriate?
n
a
n
a
a
n
a
n
n
a
a
a
n
a
a
a
a
n
a
n
a
a
a
a
a
n
a
n
a
Dose Response--
Children
• Dose response approach to be informed
by evaluation of MOA information
• When a postulated MOA is supported
for adults but not for children, a linear
low dose extrapolation will be used
as a default approach for the general
population (including children as
sensitive subpopulation)
William H. Farland, Ph.D., Director
Mailing Address:
National Center for Environmental Assessment (8601 D)
Office of Research and Development
U.S. Environmental Protection Agency
1200 Pennsylvania Avenue, NW Ariel Rios Building
Washington, DC 20460
„ -_
E-JJ
Office Location:
808 17th street, NW
Suite 400
Washington, DC 20074
Tel: 202-564-3322
FAX: 202-565-0090
E-mail: farland.william@epa.gov
-------
An EPA/NIEHS Workshop:
Information Needs to Address Children's Cancer Risk
March 30 and 31, 2000
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
Mode of Action--
Children
• When sufficient information is developed to
show a MOA that is expected to be
relevant to adults, an evaluation will be
made as to whether this MOA is relevant
to children
• When there is no information on children,
or there is no cogent rationale supporting
the comparability between responses in
children and adults, the postulated MOA
for adults is not considered applicable to
children
D
D
D
D
D
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
Upcoming Events
0 Fall 1999 - Agency and Interagency
Review
D Early 2001- Publish Final Guidelines
D Other Related Activities
+ Peer review of standard computer
algorithm (Fall 2000)
* Development of supplementary guidance
on Margin of Exposure analysis
+ Develop supplementary guidance on the
assessment of children's risk
Office Location:
808 17th street, NW
Suite 400
Washington, DC 20074
Tel: 202-564-3322
FAX: 202-565-0090
E-mail: fartand.william@epa.gov
-------
Overheads from
Frederica Perera's Comments on Topic 1: Current and Proposed Approaches to
Assessing Children's Cancer Risk
E-55
-------
-------
Table 2. Mean biomarker levels for mothers and newborns and mean differences in biomarker
levels between mother-newborn pairs among all subjects (N = 160 pairs)
Biomarker
PLASMA COTININE (ng/Hll)
PAH-DNA ADDUCTS
(per 108 nucleotides) "
PAH/ AROMATIC DNA ADDUCTS
(per 108 nucleotides) c
HPRT MUTANT
FREQUENCY (X 10"6)
Mothers
Mean ± SD (n)
8.3 ±24.5 (158)
6.4 ±9.2 (135)
14.0+14.9(139)
22.1 ±25. 6 (67)
Newborns
Mean ± SD (n)
14.0 ±35.3 (160)
7.6 ±9.6 (135)
16.0+12.2(139)
8.4 ±33.7 (64)
Mother-Newborn Pairs
P-value (n pairs) "
< 0.001 (n = 158)
0.131d(n=112)
0.002" (n= 122)
< 0.001 (n = 48)
" Difference between mother and newborn biomarker level assessed using Wilcoxon Signed Ranks Test
b Measured in white blood cells by ELISA
c Measured in white blood cells by 32P-postlabeling
d Dose of PAH to the fetus estimated be approximately one-tenth of the dose to the mother
AACR 2000
Columbia Center for Children's Environmental Health
Vulnerability of the Fetus and Child
Exposure Effect
Lead (Needleman et al.,1979) D
Radiation (SMmizu et a/., 1991) C
Pesticides (NAS, 1993) D,C
PCBs (Jacobsen andJacobsen, 1996) D
Polycyclic aromatic hydrocarbons (PAH) D,G
(Perera, Whyatt et al., 1998)
D - Developmental C - Cancer G - Genetic damage
E-57
-------
Biomarkers in Children Stratified by Ethnicity
Cotinine
(ng/ml)
A. African-American Mean 5.64
N 13
SD1 12.9
B: Hispanic Mean 1.82
N 96
SD 2.50
A>B, adjusting for ETS P=0.059
exposure
1 SD, Standard
Deviation
4-ABP-Hb
(pg/g)
26.3
4
5.08
32.8
47
16.7
p>0.1
PAH-
aibumin
(fmol/g)
.550
13
.369
.356
93
.504
P=0.021
SCE
9.80
9
1.86
9.83
55
2.34
p>0.1
TABLE 4
PAH-DNA ADDUCT LEVELS8 STRATIFIED BY CYP1A1MSPIRFLP
Placental Tissue
Infant WBC
CYP1A1 Mspl-l-
7.8 ±0.6 (125)
7.1 ±0.9 (106)
CYP1A1 Mspl+/-, +/+
11.1 ± 1.5 (30)"
9.8±2.2(28)c
a mean±SE per 10s nucleotides (n)
b beta = 0.5, < 0.01
c beta = 0.5, p = 0.06 MspI+/-,+/+ vs Mspl-/- controlling for smoking status, place of
residence, coal use, dietary PAH, home/occupational exposures.
E-58
-------
Overheads from
Topic 2: Enhanced Use of Test Data Related to Children's Cancer Risk
(Rochelle Tyl)
E-59
-------
-------
Information Needs to Address Children's Cancer Risk
Holiday Inn Arlington at Ballston
Thursday, March 30, 2000 and Friday, March 31 ,.2000
Topic 2
Enhanced Use of Test Data Related
to Children's Cancer Risk
Presented by
Rochelle W. Tyl
Research Triangle Institute
Research Triangle Park, NC
f
Reproduction Study Endpoints:
Two-generation (QPFFS 870.3800)
*,
ladices
Mating Index
Fertility Index '
Pregnancy Index
*Precoital interval m days
Gestational length in days
Prenatal (postimplantation) loss
Estrous cychcity - ^ ^
cycle length m days ^ f •>
# (%) cycling/not cycling,
abnormal cycles " -"^t
•* Stillborn Index
\ Etve birth Index
^.Survival Indices
^ %
PND 0-4 (precull)
~PND4-7(postcull)
r PND 7-14 (postcull)
* PND 14-21 (postcull)
; 'lactational (PND 4-21;
4 postcull)
E-61
-------
Information Needs to Address Children's Cancer Risk
Holiday Inn Arlington at Ballston
Thursday, March 30, 2000 and Friday, March 31, 2000
Reproduction Study Endpoints-2
Offspring
Sex ratio at PND 0,4,7T14,21
# Males, females/litter
Pup body weight by sex/litter at PND 0,4,7,14, 21
Onset of puberty:
Age at acquisition of vaginal opening
(VOXfemales), * ~
Weight atacquisition (ANCOVA)
Age at acquisitionjof preputial separation (males)
Weight at acquisition (ANCOVA)
V.
* Age at first estru& (PN^D age & time from VO in
days • ^i , '! *
" f Bf !,>•>-
Anogenital distancaand body weight by
individual pup on END 0 (triggered in F2 if
developmental/reproductive effects present in Fl)
oints-3
Reprpdiip
*
FQ. Fl.
gSWS^^^^Vi ll'"" -;' . ' ,
seminal
glands, prostate,
oviducts, uterus plus
; eeEwp|^^^c ejsmis at necropsy
' Bcift^^Sl^rV Sidneys, spleen, brain
'* :-KQik!^fff^^^i> ^ < , . x
(plus taiget|)rgans, gross lesions)
•": ^;*?««SS|sS@P^*Vv^ '
Weia^ngisiluiMb 3/sex/litter)
, spleen, target organs
E-62
-------
Information Needs to Address Children's Cancer Risk
Holiday Inn Arlington at Ballston
Thursday, March 30, 2000 and Friday, March 31, 2000
Reproduction Study Endpoints-4
Necropsy-2
Histopathology >
Parental animals
Males: 10/group: high dose anS control
one testis (GMA plastic, PAS/H)
all remaining organs (paraffin, H&E)
Males: unsuccessful breeders
H. i, i,^ ^ ^.
Females: 10/group: high dose and control
ovaries^ step-'sections-namber of primordial
follicles „ 7<- *r v
all remaining organs
Females: unsuccessM breeders
E-63
-------
Information Needs to Address Children's Cancer Risk
Holiday Inn Arlington at Ballston
Thursday, March 30, 2000 and Friday, March 31, 2000
Prenatal Developmental Toxicity Endpoints
(OPPTS 8703700)
Exposure duration: GDO/6 to term
Maternal parameters: ' -
Mortality (#,%)
Body weight, weight change
• Feed consumption (g/day; g/kg/day)
Gestational parameters:
# ovarian corpora lutea/dam or doe
# total (resorptions^dead fetuses, live fetuses)
implants/litter .
% preimplantation loss/litter >
% postimplantation loss/litter
# (%) nojn-live (resorptions plus dead
fetuses) implants/litter
# (%> affectedYnon-Iive plus malformed)
implants/litter v ' } *' ,
,
Prenatal Deyelpppr "
_-- . ;
Offspring •••-;-V;MgS
ti >^w-v *»i -•. 'Ji-T^*!r.T:t*.«-i.i-is**«?r*i;Dt»
#EelusfeS::i«^
and v-ari^ion||^^|
Post?- fio^tViB^atnilia^i
cardiqyascfilarp^^g
JiiptSSR.',^ "
i^aBSefmalfcirmations
iS'Jwsttif"^^..'"'.!?jali1-',;.^"' "'
P^^PK'--:,,
||lEisifea^eig^.all.
|pg|i^^a|Kjrmations/
^^£t&&f^-^' '• • \'
|^S|fe^tftVg¥ ;.?.>
E-64 .
-------
Information Needs to Address Children's Cancer Risk
Holiday Inn Arlington at Ballston
Thursday, March 30, 2000 and Friday, March 31, 2000
Combined Chronic/Carcinogenicity Study
(OPPTS 870.4300)
Duration of exposure. 18 months (mouse); 24 months (rat)
Age at onset of study Young adult (6-8 weeks)
Dose groups: Control and at feast three treatmentgroups, including
MTDandNO(A)EL
Evaluation times: 12 months and termination of study
In-life parameters evaluated: , "' ^
Clinical/functional observations /_ f
Ophthalmological examination, ™ * f
Body weight^weight gauiu „ ^ ^ -^ :.
Food/water consumption *• *- *
"-iv .; ,
Hematology, clinical chemistry, unnalysis •:
.'<•*.'*& *3*. *
Combined Chronfc/Carcin©^enicity Study-2
Gross necropsy ^ ^l
Organ wet weights: -T -1*
Liver, kidney ,^o!renals; testes, epididymides,
o vanes, uteras; spleen,"nbranv heart, lungs (if
inhalation sftidy) v " f »T^ / _
Gross observations: !v
.
'
,,
Tissues nojed above, jjlu&"bthers from digestive,
nervous, glandular, respttatoiy/bardiovascular/
hematopoietic, urogenitaj sjfsteins, all gross lesions and
masses, skin ^^
* ]-
E-65
-------
Information Needs to Address Children's Cancer Risk
Holiday Inn Arlington at Ballston
Thursday, March 30, 2000 and Friday, March 31, 2000
Combined Chronic/Carcinogenicity Study-3
Histopathology
Full histopathology on organs and tissues noted above-
on all control and high dose animals and all
animals dying or killed during study
on animals in other dose groups, as appropriate
All gross lesions in all animals
Target organs in all animals
- * 1
90-Day Toxicity Study in Rodents
(OPFfef£()\3100)
Duration of exposure: 90 djys^lS^weeks)
Age at onset of study: Young\a|u|f^6-8 \yeeks)
Dose groups: Control and atleasftfiree treatment groups,
including MTD and NO(X)EL
Evaluation time: at 90 days I ^j^ v<
In-life parameters evaluated,' gross necropsy and
histopathology requirements^are same as for chrome
component of 870.4300 I<|""' °*
E-66
-------
Information Needs to Address Children's Cancer Risk
Holiday Inn Arlington at Ballston
Thursday, March 30, 2000 and Friday, March 31, 2000
Immunotoxicity
(OPPTS 870.7800)
Duration of exposure: >28 days (may be incorporated in
90-day or chronic study)
Age at onset of study: Young adult rodent (6-8 weeks)
Dose groups: Control and at least three treatment groups,
including MTD and NO(A)EL F
Evaluation time: at >28 days
JU- ^ -* >-
In-life parameters evaluated: f" t
Clinical observations, body weight, food/water consumption
Immunotoxicity functional tests: Either the antibody plaque-forming
cell (PFC) assay or immunoglobulin quantification using the enzyme-
linked immunosorbent assay (ELIZA) ,-
^ „ \ F ^ ~
Phenotypic analysis of total B ceU/jtotal T cell, T cell subpopulations
from spleen or peripheral blood by flow cytometry
*
MetalpoligU^^^^g^c^iiietics
Duration
Age at
Dose groupsi
Evaluation time: iip> to@f
Parameters
'scit-^-S weeks)
•
onrpercent^p(^KeS®$admihistered dose from
eces,.'exhal^^^P^^|&^:.;
^«*wi^iK^lft^^^^S«S'i.;r''".'
.,,. .-
Tissue
unchanged testsubSiffli^lMinetabdltes in excreta
' ' ^'''' " '
E-67
-------
-------
Overheads from
Topic 3: Future Directions for Toxicology
Testing to Address Children's Cancer Risk
(Retha Newbold)
E-69
-------
-------
CO
-------
d
d
O
d
o
d
o
o
H
•
-------
cd
o>
OH
13
**
O
•PN
S
W
Q
CO
2? <+-•
£ o
+->
s g
5? S
£ *
jrt
X)
r75
^^
CO
->,-
CO
d
o
CO
s o
W ^ o
a s? s
a
CO
s ^
& a>
SL
S VH H
-J di T3
a
VH
eg
CD
•a
o
CO
D
a
Q
^
.-I
CO
d
o
U
ces
^ ^r
d
W)
2 -a
. «s
w)
<4-H
O
•H M
b
• d ^
d c3
o o
3' 'TS ^>
•s a -S;
?i
<« 8
o cL -
(D
-------
4-1
O
S
3
1
13
.$H
o
,°
J4-H
^— 1
•D
T3
O
• l-H
O
1
s
*<3
03
OH
13
00
13
•4— '
03
•S
•acterization of experimental tumors in
»— i
03
43
O
<^
G
O
e/i
Systematic comparii
msters.
03
orT
o
animal models... mi<
logenic potential of estrogens.
«
• I-H
O
•s
JHM
Focused research or
- metabolism
- DNA adducts
&
• PH
- mutagenic activ
§
. »_4
- cell transformat
& animal tissues
c
of huma
00
• I-H
00
4^
13
§
O
T5
0
1
G
^
1
43
(D
00
VH
4^
|
00
•4—>
1
oo
• 1-H.
CO"
0>
42
0
03
S
03
4S
'S c^
O G
d carcino
carcinoe
chemicals are indee
with other chemical
E-74
-------
E-75
-------
00
00
^2
A
fi
d^
M^
^<
PH
CM
0
C£
ts
^
ta
^
3
^S
O)
s
Pu
^0
"33
>
o>
O
Comparative
C/2
fi
C3
S
ffi
HH
^d
0)
O
•P^
^
fi
•^
C^J
a
o
^4^
^1 "V
V
fc-
s
VI
O
ft
^
«
bfl
1
O
•1
S
1 i
+m*
«^H
• 1-^
i
H-H
G
• i-H
^
4— >
•f-H
i
.
C
P
0
O
1
OH
00
00
cd
o
0)
Q
•
"O
G
cd
i— i
rrt
Oviductal, cervici
4— >
• i-H
o
^
s
o<
o
.y
oo
'fi.
o
OH
X
•
oo
G
• 2
uterine malformal
oo
4— »
oo
o
1
• i-H
•f-H
OH
•
ft .
^f"H
O
Paraovarian cysts
00
VH
o
S
3
^
3
• Sn
*4— >
00
H
•
G
• f-H
mesonephric orig
o
•i-H
*niniz
,
S P^
• •
G
^^
O
»r1
•^— ^
O
G
| «§
-§ >»
• in T3
§ S
00
2
o<
•
^^j
G
Vaginal adenosis
adenocarcinoma
E-76
-------
03
S
C
-------
a
e
CO
Q
JBt
•^^
d
o ,
ft
ji
«
o
to
X
01
•*->
3
r*
t
in
E-78
-------
Adenocarcinoma of ihe Uterus
E-79
Neonatal DES treatment (2fig/pup/day on days 1-5) induces uterine
adenocarcinoma in 95% of the mice that are 18 months old or older.
-------
C
o
c:
o
o>
V3
s<
l^»
o
S
»ti
-4-^
«t-i
O
^*->
S3
0)
s
a-
.2
"3
>•
a>
"O
"93
-*^
93
• C
0
&>
ti
b£
afi
s
T3
C
o
Estrogen administrati
.E3
93
£
o
.S
"3
u
93
o
O
fi
o>
TJ
93
«M
O
Q
O
C
^
"3
s
t-
&
bf
IS
93
*^
T3 .
4) O
-*^ s?
93 0s
"3-2
o ^
W5 A,
W5 S-^
93 o>
-gg
5 tS
V)
• IH^
cs
V3
«
fi
O
4=
^
bX)
c
•p«<
V3
tM
dependeni
fl
bX)
o
SM
Uterine tumors are es
A \
QJ
U
S
O
.0
>>
S
o
-*•*
o
«
»•*
u
93
K^
O
• e\
"O
4>
T3
T3
93
.2
fi
c
P0
^
bX)
C2
withdrawn and growii
'S
i
c-
13
4^
'S
puberty inhibits tumo
c
o
"O
S3
93
0C)
-S
c
o
L-
£
O
2
is
%«*
d
'o
&
C/3
As evidence of neopla
"93
c
««•*
£
o
T3
JD
93
0>
J3
•*^
«*-i
O
O
"O
'5
o
0>
N
• ••^
V5
w*
ten metasti
'o
^j
o
c
o
o
1
H
•
93
bX) ^
93 £>
<^j> "^
O 93
00 0
fi
*c
4>
H-^
0
^g
i
IHg
S3
W5
O
fc.
O
V5
"93
J2
13
w
*S
oj
js
0
estrogenic
i*
^
•*•*
o
Neonatal exposure to
adenocarinoma.
E-80
-------
A
O
4?
0)
03
O
P V
O
O
E-81
-------
5^
p^
cd
m
cn
en
en
=5
O
S
cs
^2 ^N
CQ F*N
S §j
d o
.22 '
'"" C3
«
(DD P-I
^ 0
0 t?
^_ M
OJO o>
PM p
a) rs
o S
0)
£ wo
pt; fl
rt o
^ fl
? c«
d
o-2
S fl
DJO S
CS CM
en
PM P _
CJ PH O
OX) en gj
"I -S ^
3 ^i .5
en
-2 .2 .2
-x 'M Z*
C/3
«
- s
a g g
ft1 S o
M
= -i
fl
o
3
en
O
fi —
S. -g
& CM
O
en
s «
fi §
a;
S S+-1
O CM
en
OH
X
en J2
^ •» '*••
SH
^
tC W0
^-5 CM
S o
en
O
cu
1
fl
?M o
4<—' PM
O *3
en
2 .tJ
2 «
e ^-*
£3 ea
fl
o
en
en
Q ^
PM g
CC en
Cr*
M
a> c:
.§•§
en
fi
O
en
.2
C5
H SP
E-82
-------
,g
I
CC
05
Q
.so
^
c«
SiH
«w
«)
C/3
S
O
CO
S
C3
S
P"N
03
PN
^
t3
s
c
a,'
OD
X
.0
hN
i-M
3
G
cd
G
•PN
O
o
CJ
PN
13
• PN
o> 'cS
ft
o
G
CS
ft E3
OS ^
s ^
s ^
O) PN
« 'C
«
-------
s
H .0
13
O) fi
S
1/2
=
CS
=
cs
E-84
-------
E-85
-------
s
C/D
,£f
'35
Q
3
ts
O u
C/3
3
O
1
o
a o
• i-H *-H
0)
i<
•
E-86
-------
I
00
Q
^
o
o
1 i
0)
O
00
00
CJ
0)
OH
X
s
o\
8 «
e 2
t *
T «
Newboldj
o
H
2
H
' w
Si
'•3 is
s- ft
O
•*^
S
K
Q
i o
U
O
W3 =
Q
\ =
X
a
E-87
-------
g
S
0
C/)
o
IDD
.3
d
02
<
•s*
^^ T^S mf
d d S
cj d
r£ ' 8
flJ
s-?
Gfl
33 I
••W ^^^n H^^
Q Q
o>
"o
(d'
c»
a
a
=
0
e
0)
d
E-88
-------
•s
C/3
a
£
C/5
QJ
animal models to screen for adverse
*o
CQ
en
O
o>
Further developed mi
re to DES and other environmental
§9
O
13
'S
effects following pren
•8
a
1
I*
o
0
endocrine disrupting
exposure on other organ systems
en
Q
©
s
- transgenics
- DES-lineage
Assessed effects of in
|
en
- immature
- skeletal
- endocrine
- central nervous s
•s & redefined classical hormone recep
**
£
cu
8
i*
G
O
1
O
^
^ CQ
0) pC
fl ja.
T3 93
4) ^^
•^3 jsj
C i
o>
T3
HH
- ER beta
i
en
•g
CQ
0
Epigenetic/genetic me
Molecular markers
E-89
-------
-------
Overheads from
David Wallinga's Comments on Topic 3: Future Directions for Toxicology
Testing to Address Children's Cancer Risk
E-91
-------
-------
Toxicology Testing & Children's
Cancer Risk
David Wallinga, M.D.
Public Health Program
Natural Resources Defense Council
March 31, 2000
Cancer Assessment is Data Poor
> 80,000 registered industrial chemicals
> 2,863 HPV chemicals produced at
greater than 1 million pounds per year
> For 75%, EPA has no developmental
toxicity information at all (Goldman 1999)
63% of high volume chemicals lack any
test of carcinogenicity (EOF 199?)
523 pesticides registered in food/feed
Natural Resources Defense Council
David Wallinga, M.D.
E-93
-------
National Research Council, 1993
"In general, the committee found that current and past
studies conducted by pesticide manufacturers are
designed primarily to assess pesticide toxicity in sexually
mature animals. Only a minority of testing protocols have
supported extrapolation to infant and adolescent animals.
Current testing protocols do not, for the most part,
adequately address the toxicity and metabolism of
pesticides in neonates and adolescent animals or the
effects of exposure during early developmental stages and
their sequelae in later life. (Pesticides in the Diets of
Infants & Children, p. 4)
Natural Resources Defense Council 3
NRC & Children* s Susceptibility
• The NRC (1993) marshaled ample scientific
data suggesting that infants and children,
generally, have greater susceptibility to the
toxic effects of chemicals....
• .. .Even if complete testing for individual
chemicals to characterize these risks to
children is usually lacking.
Natural Resources Defense Council
David Wallinga, M.D.
E-94
-------
Sources of Children's
Susceptibility to Carcinogens
> Differences in inhalation & absorption.
1 Differences in activation and detoxification, combined
with relative immaturity of protective mechanisms,
such as immune system and blood-brain barrier.
Inherent vulnerability of rapidly dividing and growing
cells, as well as developing organs.
Differences in diet and behavior.
Natural Resources Defense Council
Importance of Strong Defaults
' :ra:*s*R:i*BaBiBi1^^
A child's general susceptibility to toxic
chemicals, in the face of data gaps on
individual chemicals, leads to the need
for...
• Strong, health-protective default
assumptions in cancer risk assessment
• High hurdles for abandoning these defaults
Natural Resources Defense Council
David Wallinga, M.D.
E-95
-------
Implications for Cancer Risk
Assessment Today
• ----^--^SECT»g»'»»^"™*»'°" •""'"L"m"""'"" "' "'"
Default assumptions in the proposed
Cancer Risk Assessment Guidelines
need to be closely examined
Claims that current defaults are health-
protective should be backed up by data
~ Example: the Guidelines currently
have no default for human variability.
Natural Resources Defense Council
Implications for Future Testing
^^^•••••••••••••••••i J J!U'"_^^££S--g,
• Extend developmental toxicity tests beyond birth, to
account for latency.
• Assess cancer risk from pre-conceptual exposure.
• Look at effects of short-term carcinogen exposure
during developmental windows.
• Require pharmacokinetics in immature animals.
• Build developmental windows of vulnerability into the
testing paradigm.
• Validate & test for endocrine disruption.
• Do semiquantitative assessments of cumulative risk
Natural Resources Defense Council 8
David Wallinga, M.D.
E-96
-------
Pharmacokinetics
"Critical to the interpretation and extrapolation of data
on developmental toxicity is an understanding of the
Pharmacokinetics of chemicals in the developing
system and the complexities of direct and indirect
developmental exposures during pregnancy, lactation,
and to neonates by various routes of
exposure... .Guidelines for appropriate
pharmacokinetic information relevant to pre- and
postnatal exposures are needed."
-- Report of the EPA Toxicology Working Group of the 10X Task
Force, April 1 999.
Natural Resources Defense Council 9
Developmental Carcinogenesis
' • •^^•'•••••••^•••^••••••••^•••••l tULLBOBSSK V. v. ••
"Likewise, no specific standardized guidelines
exist for second tier testing on chemicals that
are suspected to enhance carcinogenic
response following perinatal exposure.
Experience with testing agents for
carcinogenic potential following exposure
during development is very limited to date."
Report of the EPA Toxicology Working Group of the 10X Task
Force, April 1999.
Natural Resources Defense Council
10
David Wallinga, M.D.
E-97
-------
Endocrine Disrupter Screening
Mandated under both SDWA and FQPA
No validated screen as yet.
Modes-of-action detected by these screening
tests will suggest a heightened concern for
risks of exposures during developmental
periods.
Highly relevant to controversial modes of
action in endocrine-active organs: thyroid,
mammary gland, adrenal etc.
Natural Resources Defense Council
11
Protecting Children Can't Wait
• Proposed Guidelines have been applied to at least 45
pesticides since August 1996.
«> Draft Guidelines used to revise IRIS or create new IRIS files
for 15 chemicals from 1996 to 1998, including arsenic,
benzene, chromium IV and PCBs.
o New or revised IRIS assessments for another 22 chemicals
anticipated in FY2000, including 1,3 Butadiene, cadmium,
chloroform, formaldehyde, and toxaphene.
e New or revised IRIS assessments anticipated for another 27
chemicals in FY 2001, including acetaldehyde,
benzo(a)pyrene, and ethylene oxide.
Natural Resources Defense Council
12
David Wallinga, M.D.
E-98
-------
Overheads from
Topic 4: Epidemiologic/Molecular Epidemiology Information to Address
Children's Cancer Risk
(Leslie Robison)
E-99
-------
-------
Epidemiology/Molecular Information to
Address Children's Cancer Risk
• Relatively Rare
• Unique from adult malignancies
• Increasingly recognized as biologically distinct
subgroups
• Relative lack of etiologic information
IS THERE A NEED FORA NATIONAL INITIATIVE?
National Network for Research
on Cancer in Children
A national registry of children with cancer for identifying
environmental & other causes of childhood cancer
To build upon the unique national clinical trials system for
treating children with cancer
To identify children at the time of their cancer diagnosis, allowing
collection of tissue specimens
To coordinate efforts with population-based cancer registries
To support and facilitate scientific studies of the highest merit by
the qualified investigators to study causes of cancer in children
E-101
-------
Why a "National" Network to Study
Causes of Cancer in Children ?
The cancers of children differ from those in adults
in their histology and in their biological
characteristics
8700 new cases diagnosed annually in children
younger than 15 years of age, and the numbers
of children with any given diagnosis is relatively
small
Why a "National" Network to Study
Causes of Cancer in Children?
Within many tumor diagnoses are subtypes defined
by specific chromosomal changes:
• these distinctive subtypes are likely to arise through
different pathogenetic mechanisms
A National Network allows identification of sufficient
numbers of cases to permit study of distinctive
biologically-defined categories of tumors defined by
specific alterations in tumor-associated genes
The National Network needs to have the capability of
defining tumors by their key biological characteristics
E-102
-------
Estimated Numbers of Cancers
in Children in U.S.
Ewlng's sarcoma
Osteosarcoma
Wilms
Neurobtastoma
Acute myelold
leukemia
Acute Lymphoid
Leukemia
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200
Estimated Number of Cases in U.S. per Year
Estimated Number of Cases of ALL Annually in
U.S. by Molecular Subtype
TCR and TAL1
ABL-BCR: t(9;22)
PBX1-E2A:t(1;19)
MLL: t(11;v)
TEL-AML1: t(12;21)
Hyperdiploid
50 100 150 200 250 300 350 400
Estimated Number of Cases in U.S. per Year
E-103
-------
Estimated Numbers of Brain Cancers
in Children in U.S.
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Estimated Number of Cases in U.S. per Year
Why a "National" Network to Study
Causes of Cancer in Children?
i The causes of most cancers in children are not
known.
i Known Risk Factors for Childhood Cancer
explain only a small percentage of cases:
• Prenatal diagnostic x-ray exposure
• Postnatal therapeutic radiation or chemotherapy
• Down syndrome, and other genetic diseases (e.g.,
Bloom syndrome, neurofibromatosis, ataxia
telangiectasia)
E-104
-------
Factors with Inconsistent or Limited
Evidence Linking to Childhood Cancer
Maternal and paternal
smoking prior to and during
pregnancy :
Parental occupations and
occupational exposures
Diet
Electromagnetic fields
Indoor radon exposure
Vitamin K prophylaxis in
newborns
ii Parental and child
exposure to pesticides
n Maternal alcohol
consumption during
pregnancy
•i In utero and postnatal
infections
• Other environmental toxic
exposures
Why a "National" Network to Study
Causes of Cancer in Children ?
Opportunities Justifying National Network
• Advances in molecular characterization of
tumors
• Advances in exposure assessment
methodology
•Advances in understanding and evaluating for
genetic susceptibility
E-105
-------
Improved Methodologies/Technologies for
Causation Research
• GIS technologies for hypothesis generation and
correlation
• Sophisticated categorization of occupational
categories relevant to specific exposures
• Exposure assessment (ability to detect minute
quantities of substances in biological fluids and in the
environment)
•i Identification of biological markers of exposure and
susceptibility
•i Identification of potential genetic susceptibility factors
Why a "National" Network to Study
Causes of Cancer in Children?
To Overcome Limitations of Previous
Childhood Cancer Causation Studies
• Insufficient numbers to identify meaningful
increases in risk
• Analyses combining diverse, biologically distinctive
tumor types
• Inadequate/unreliable assessments of exposure
Secondary benefits: Studying patterns of care,
enhancing national surveillance capabilities
E-106
-------
Pediatric Clinical Trials Cooperative Groups
m Children's Cancer Group and Pediatric
Oncology Group represent over 200
institutions throughout U.S. and Canada that
are involved in the treatment of most children
with cancer
m Register approximately 94 percent of children
under the age of 15 years diagnosed with
cancer
m Unifying into single Pediatric Clinical Trials
Group: The Children's Oncology Group
CCG Member Institutions
in North America and Australia
O Full Member
° Affiliate Member
CCG
Beyer
E-107
-------
Pediatric Oncology Group Institutions
(Continental United States)
Childhood Cancer Epidemiological Studies
Conducted Through Cooperative Groups
Osteosarcoma
Heptatoblastoma
Swing's Sarcoma
AML
Wilms' Tumor
Retinoblastoma
NHL
Infant Leukemia
Twin Concordance
PNET and Astrocytoma
Hodgkin's Disease
ALL
Neuroblastoma
HIV- Malignancies
PNET/Medulloblastoma
Rhabdomyosarcoma
E-108
-------
Opportunities Created by a Single
Pediatric Clinical Trials Group
\ Most children with cancer in U.S. will be seen
by COG member institutions
Institutions will request patient/family
participation in the Network at or near time of
diagnosis
Opportunity for collection of tissue specimens
(e:g., tumor tissue, blood, buccal cells) at time
of diagnosis
Key Elements of National Network
i At time of diagnosis at GOG institution,
permission requested to include the patient in the
Network's registry of childhood cancer cases.
• Family consent to be contacted to consider participation
in future studies relating to identifying causes of
childhood cancer.
After permission is granted, information
concerning the case is submitted to the COG
Data Center
E-109
-------
COG Registration/Consent Protocol
'Level" of Registration
COG Case Registration
\
Consent to Register with Identifiers
Consent to Be Contacted
\
National
Network
Registration/Consent Task Force
• Registration/Consent Protocol
• Informatics
• Biological Specimens
«> Collaboration with State and regional registries
« Assess institutional practices and resources
• Cancer registration consultants
« Research infrastructure
E-110
-------
COG Registration/Consent Task Force
Year 2000 Timeline
Registration/Consent Protocol
Informatics
Biological Specimens
Collaboration with State and regional registries
Assess institutional practices and resources
Cancer registration consultants
Research infrastructure
Fall
Summer
Summer
Spring
Spring
Spring
Summer
Key Elements of National Network
i Serves as a resource for scientists wishing to
conduct hypothesis-driven research studies
relating to potential environmental and/or
genetic causes of cancer in children.
Network responsible for establishing Committee
to review proposals to utilize the Network.
Review based on:
• the proposals' scientific soundness and merit, and
• on inclusion of appropriate safeguards for
participating patients/families
E-lll
-------
PEDIATRIC CANCER SURVIVORS
Precise Exposure
Assessment
Defined Mechanisms
Access to subjects
Monitor Progression
Influence of primary
disease
i disease
Medical Exposure
*-—- 1 Relevance of Mechanism
I of Action
1 Route of exposure
Intermediate
Markers
Range of exposure
Biomarkers of
Effect/Progression
T
Well characterized | Cancer Outcome |
STUDY OF SECONDARY LEUKEMIA
Alkylating Agents Genetic susceptibility
Medical Exposure [
Intermediate
Markers
Biomarkers of
Effect/Progression
Exposure markers
Intermediate markers
Cancer Outcome | Markersof effect
T-MDS, AML
E-112
-------
Overheads from
Workshop Summary
(George Lucier)
E-113
-------
-------
X
e «
O
to
U-i
Ui
b-
2
Ui O
z, ^
^ <0
—* ^T
\U X
Q Z.
•^^^n^v
>
dj
&
P
UJ
fe£
UJ
5
UJ ^
IU
UJ
0§
W OC
HI ^
UI Q
« -
S|
E-115
-------
• 4
2
O
CD
Z.
LU
UJ
O
$0
in Q.
_i 50
Q
10
Z Cs:
ill (X
DEVELO
O
o
<^r>
^P<
> ^"w c L r— «W5>
su p rvvk te^
£iSO
fexl1^
3 -^^Q
O
0^
Ul
u.
LL.
-^ Q CO
tn
E-116
-------
*%
ILL
IB
£
U-I
fci
tu
<^
O
ui
8512 .j.
(JJ O «.
ouj
t-CQ
(^ (u O
E8
DO.
Q X
I
Q
= Ul
O
|
^
t<
E-117
-------
US
&
§
C£
LU
^^^
OJ
QC
o
Ul ,M
S3-
D >
Hi
LS
lu
O |5
O
-------
APPENDIX F
LIST OF BACKGROUND MATERIALS
PROVIDED TO PARTICIPANTS
PRIOR TO THE MEETING
-------
-------
Document entitled "Comparison of the effects of chemicals with combined perinatal and adult exposure vs. adult only exposure in
carcinogenesis bioassays."
Report of the 1996 FIFRA Scientific Advisory Panel meeting addressing "Comparison of the effects of chemicals with combined
perinatal and adult exposure vs. adult only exposure in carcinogenesis bioassays."
Document entitled "A proposed OPP policy on determining the need for in-utero/perinatal carcinogenicity testing on a pesticide."
Report of the 1997 FIFRA Scientific Advisory Panel meeting addressing "A proposed OPP policy on determining the need for in-
utero/perinatal carcinogenicity testing on a pesticide."
Background paper on availability of toxicity testing data for assessing cancer risk.
American Academy of Pediatrics, Committee on Environmental Health. Cancer. In: Handbook of Pediatric Environmental Health. Elk
Grove Village, IL: American Academy of Pediatrics.
Colt, J.S., and A. Blair. 1998. Parental occupational exposures and risk of childhood cancer. Environmental Health Perspectives
106(Supplement 3):909-925.
Legler, J.M., L.A.G. Ries, M.A. Smith, J.L. Warren, E.F. Heineman, R.S. Kaplan, and M.S. Linet. 1999. Brain and other central
nervous system cancers: Recent trends in incidence and mortality. Journal of the National Cancer Institute 91(16): 1,382-1,390.
Linet, M.S., L.A.G. Ries, M.A. Smith, R.E. Tarone, and S.S. Devesa. 1999. Cancer surveillance series: Recent trends in childhood
cancer incidence and mortality in the United States. Journal of the National Cancer Institute 91(12): 1,051-1,058.
National Research Council. 1993. Executive summary. In: Pesticides in the diets of infants and children. Washington, DC: National
Academy Press, pp. 1-12.
Perera, P.P. 1997. Environment and cancer: Who are susceptible? Science 278:1,068-1,073.
Perera, P.P., R.M. Whyatt, W. Jedrychowski, R. Rauh, D. Manchester, R.M. Santella, and R. Ottman. 1998. Recent developments in
molecular epidemiology: A study of the effects of environmental polycyclic aromatic hydrocarbons on birth outcomes in Poland.
American Journal of Epidemiology 147(3):309-314.
Perera, F.P., W. Jedrychowski, V. Rauh, and R.M. Whyatt. 1999. Molecular epidemiologic research on the effects of environmental
pollutants on the fetus. Environmental Health Perspectives 107(Supplement 3):451-460.
Ries, L.A.G., M.A. Smith, J.G. Gurney, M. Linet, T. Tamra, J.L. Young, and G.R. Bunin (eds). 1999. Cancer incidence and survival
among children and adolescents: United States SEER Program 1975-1995. National Cancer Institute, SEER Program NIH 99-4649
Bethesda, MD.
Tang, D., D. Warburton, S.R. Tannenbaum, P. Skipper, R.M. Santella, G.S. Cereijido, F.G. Crawford, and F.P. Perera. 1999.
Molecular and genetic damage from environmental tobacco smoke in young children. Cancer Epidemiology Biomarkers &
Prevention 8:427-431.
U.S. Environmental Protection Agency. 1999. Guidelines for carcinogenic risk assessment (review draft). Washington, DC.
Zahm, S.H., and M.H. Ward. 1998. Pesticides and childhood cancer. Environmental Health Perspectives 106(Supplement 3):893-908.
F-3
-------
-------
-------
United States
Environmental Protection Agency/ORD
National Center for
Environmental Assessment
Washington, D.C. 20460
Please make all/iecessary changes on the below label,
detach or copy, and return to the address in the upper
left-hand corner.
If you do not wish to receive these reports CHECK HERED ;
detach, or copy this cover, and return to the address in the
upper left-hand corner.
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
Official Business
Penalty for Private Use
S300
EPA/600/R-00/105
------- |