Chemical Contaminants in Breast Milk:
Impacts on Children's Health 
Conference in NYC 5oct01

Organized by the Mount Sinai Center for Children's Health
and the Environment on at the New York Academy of Medicine in New York City.

• Meeting Notes

• Abstracts

Agenda

8:00 a.m.

Registration and Breakfast

8:30 a.m.

Welcome and Overview
Michael McCally, M.D., Ph.D., Mount Sinai Center for Children’s Health and the Environment

8:35 a.m.

Morning Keynote Speaker
Kenneth Olden, Ph.D., NIEHS

9:15 a.m.

Charge to the Workshop
Babasaheb Sonawane, Ph.D., U.S. Environmental Protection Agency
Donald Mattison, M.D., March of Dimes Birth Defects Foundation

9:30 a.m.

SESSION I. Patterns of Infants’ Exposure to Chemicals in Breast Milk: Analytical Considerations
Moderator: Michael McCally, M.D., Ph.D., Mount Sinai Center for Children’s Health and the Environment
Speaker: Larry Needham, Ph.D., Centers for Disease Control and Prevention
Speaker: Judy LaKind, Ph.D., LaKind Associates, LLC
Speaker: Matthew Lorber, Ph.D., NCEA, EPA
Discussant: Gina Solomon, M.D., MPH Natural Resources Defense Council

10:30 a.m.

Session I Discussion

10:50 a.m.

Break

11:10 a.m.

SESSION II. Pharmacokinetics of Toxic Chemicals in Breast Milk
Moderator: Joel Forman, M.D., Mount Sinai School of Medicine
Speaker: Jeff Gearhart, Ph.D., Wright-Patterson AFB, Dayton, OH
Speaker: Rebecca Clewell, M.S., Wright-Patterson AFB, Dayton, OH
Discussant: Elaine Faustman, Ph.D., University of Washington

12:00 noon.

Session II Discussion

12:20 p.m..

Luncheon Presentation. Global Perspectives in Breast Milk Contamination: Toxic versus Infectious Hazards
Jenny Pronczuk, M.D., World Health Organization

1:50 p.m.

SESSION III. Assessing Outcomes in Children’s Exposure to Toxic Chemicals in Breast Milk
Moderator: Philip Landrigan, M.D., M.Sc., Mount Sinai School of Medicine
Speaker: Walter Rogan, M.D., National Institute of Environmental Health Sciences
Speaker: Peter Scheidt, M.D., M.P.H., National Institute of Child Health and Human Development

3:00 p.m.

Session III Discussion

3:20 p.m.

SESSION IV: Research Needs and Implications for Risk Assessment
Moderator: Philip J. Landrigan, M.D., M.Sc., Mount Sinai School of Medicine
Speaker: Kim Hooper, Ph.D., California Environmental Protection Agency
Discussant: George Daston, Ph.D., Procter &Gamble
Discussant: Sherry Selevan, Ph.D., NCEA, EPA

4:15 p.m.

Session IV Discussion

4:45 p.m.

Concluding Remarks: Vision for the Future

Michael Lerner, Ph.D., Commonweal

Meeting Notes

1. McCally:

Thanks to: Sonawane, March of Dimes (MOD), ATSDR, NICHD, CDC, NIEHS, etc. Welcome to NYC. Moment of silence for World Trade Center disaster.

2. Landrigan:

Introduced Ken Olden, director of NIEHS

1. Ken Olden: Thanks to MSSM, Landrigan, Selikoff.

NIEHS is successful because it reaches out to the public. Children are at the top of the NIEHS agenda. Issues like ADHD are important. Along with NIEHS, established the longitudinal cohort study, with the help of PEW, Landrigan, and Lynn Goldman. The U.S. lacks a tracking system. This system should start at preconception, and follow to the remainder of their lives. The longitudinal study stops at age 20.

Strategies/Priorities: For childrens health: Develop research activities to investigate the interactions of stage of development, behavior, and susceptibility of children to environmental exposures. There are now 12 such centers. Children don’t lobby, they were forgotten for many years.

Toxicology in post genome era: Toxicology has had a slow evolution. It was limited by methods and technologies. There are complex interactions involved in disease (such as in birth defects, cancer, Parkinson’s disease). Toxicology has traditionally studied the effects of one chemical, one effect at a time. We can now look at multiple interacting pathways. The new science of Toxicogenomics will unshackle environmental health and toxicology. Toxicogenomics is poised to play a leading role in biomedical research, in the debate between environment and genetics. Nothing did more for the field than cloning the genome. (See Lichtenstein et al., paper on the attribution of cancer due to environmental vs. genetic causes, 2/3 of cancer is environmental or the product of gene-environment interaction). There is a need to address both environmental and genomic health effects. Toxicogenomics can now look at complex gene-environment interactions. You need to understand cancer pathogenesis in order for prevention efforts to succeed. Toxicology is important for identifying environmental factors that may cause harm, and for identifying harmful chemicals in human tissue. According to Eric Lander, Director of MIT Center for Genomic Research "Genes are only a small part of our makeup. The environment has a spectacular impact" (9/25/01). Individual susceptibility is controlled by many factors: age, race, genes, gender, nutrition, health status, and other environmental factors. Genes express themselves under specific environmental situations. (Diagram of human health/disease in center of triangle, with environmental exposures, intrinsic genetic susceptibility, and age/time at each side of triangle. This conference focuses on the age/time component of the triangle, exposure in utero, during crucial times of development). Genes may "load the dice" in favor or against the development of a specific phenotype, but do not dictate when or if the specific phenotype will be expressed.

Impact of Genomics on Toxicology: 1. The development of new and more informative test systems for toxicity or carcinogenicity assessment. 2. The elucidation of the genetic basis for differences in susceptibility or response to drugs or other environmental xenobiotics. 3. The understanding of mechanisms for metabolic pathways affected by toxin exposure (important for risk assessment). 4. The development of biomarkers for use in quantifying exposure. 5. The development of tools and resources for use in population-based studies.

Toxicogenomicsà Proteomicsà Metabonomics

Grand goal of Toxicology: characterize entire genotype in terms of proteomics and metabonomics, then we’ll be able to do automated carcinogenecity testing in a matter of days.

The Search for Susceptibility Genes

1. Environmental Genome Project (candidate gene approach)

Science, 278:569-570, 1997

Nature Genetics, 18:91-93, 1998

Nature Reviews Genetics, 1:149-153, 2000

B. Single Nucleotide Polymorphism Project (a trans-NIH project)

Science, 278: 1580-1581, 1997

C. New Initiatives: the National Center for Toxicogenomics, the Comparative Mouse

Genome Centers, the Environmental Genome Project

Summary: Toxicogenomics is a toolbox of new technologies that can be used to address some of the intractable problems that have long characterized the field of Toxicology. Hypothesis driven research will proceed from the application of these new tools. We will be able to examine the exposure-disease relationship from the perspective of the population.

1. Babasaheb Sonawane: Topics of conference: what we know about exposure, chemicals, kinetics of elimination, body burdens, mother’s milk, risk assessment.
2. Don Mattison: Has been surprised by questions asked by parents. They force us to reconsider questions. Toxicology is based on the use of animal models in the predictive/preventive mode. We need to reexamine issues.

1. Larry Needham: Explained the chemicals selected for the CDC Exposure Report. Persistant Organic Pollutants: dioxins/furans, PCB’s, Organochlorines. M Organometallics: methylmercury, tributyl tin. Non-Persistent Compounds: pesticides, personal care products and additives (phthalates, polybromated diphenyl ethers, fluorinated sulfonates and other surfactants, i.e. Scotchguard).

Factors to Consider when Assessing Exposure: donor selection, questionnaires, collecting and storing samples (use of preservatives), sample pools, transportation, analyses (consistency by using inter-laboratory studies), and results reporting.

There is an International Meeting of Exposure Assessment Scientists in Charleston during November.

2. Judy LaKind: Reporting from papers written for the Technical Workshop: Breast Milk Monitoring for Environmental Chemicals.
   From Cheston M. Berlin Jr., Objectives of a Breast Milk Monitoring Program:

Use of women from diverse geographic regions, different socioeconomic and demographic backgrounds; extend previous studies; increase the number of environmental chemicals analyzed; collect longitudinal information during the course of lactation; monitor any changes in the concentration of chemicals over time in breast milk; make results available (with interpretations) in peer-reviewed literature and public sources (over the internet); promote the harmonization of sampling and analysis protocols and improve the comparability of results; improved basic risk analysis (risks and benefits of breast feeding vs. formula); and improved risk communication. An important question is what to do if chemical contamination is detected.

3. Matthew Lorber: TEQ’s = Toxic Equivalent Quotients, a first order pharmacokinetic model. The half life is low at infancy, peaks around five years of age. For further information, see Abraham (1995, 1998, 2000), and Kreuzer (1997).

There is not much data from the blood of infants. One conclusion determine is that breast fed infants have at least an order of magnitude higher concentration of dioxin-like compounds (see Abraham, 2000, for infant-mother data sets). In these data sets, blood measurement is taken of infants at one year (the dependent term), and mother’s milk concentration provides the independent source term. Finally, toxins dissipate much more rapidly in infants.

Findings: Initial TEQ concentration in mother’s milk is 25 parts per trillion (ppt). This concentration decreases by fifty percent in the first six months of breast feeding, followed by a further decline of fifty percent by twelve months after birth. At birth, the average daily dose is 242ppt, going to 18ppt by twelve months. The average breast milk concentration is 87ppt during the first year. This compares to the average adult daily dose of 1 picogram. Therefore, infants average 87 times the adult dose of dioxin-like compounds during the first year of breastfeeding. Breastfed infants have higher lifetime concentrations than formula fed infants.

Conclusion: The peak body burden of dioxin-like compounds in infants (40-50ppt) occurs at two months of age. Breast milk intake of dioxins appears significant. The body burden in breast fed infants is ten times that in formula fed infants.

4. Gina Solomon (Discussant): There is patchy information in this area of research. Four main problems: A. The data comes from a small number of countries, the U.S. has produced little research in this area. B. The data is mostly cross sectional, making it hard to draw inferences regarding time trends. C. The range of chemicals tested is narrow. D. There is a lack of comparability between studies (Especially with PCB data). There are differences in sampling over time, analyzing first lactation or subsequent lactations (more comparability problems)

Needs: A. Need to "connect the dots". This is accomplished by testing mothers, breast milk, and infants. B. Avoid sample pooling (information is lost by pooling data sources). C. Establish a time period for measuring (measure during early phases of lactation, 2 weeks preferable over 8 weeks).

Questions: Landrigan: There is a wide range of dioxin values in the Abraham paper. We need to get away from doing average exposures in risk analysis. We should be reporting ranges of values instead. Kim Hooper: You can have uniform exposures but different concentrations (from variations in processing, metabolism). Lorber: Non-metabolic loss in infants overwhelms metabolic loss. Ricky Perrera: Researchers at Columbia are conducting a prospective cohort study of over 400 women and over 400 newborns through age five. It measures urine biomarkers, maternal blood, and umbilical cord blood. Larry Needham: We need a study to compare maternal serum and breast milk.

George Daston: There is data on dioxin exposure in animals. Effect markers exist, such as ones for changes in thymus-mediated immunity. TEQ ratios are set at one point in the dose-response curve. This assumes a constant toxicity ratio (poor assumption). There probably is variability in TEF’s. Lorber: The key metric in exposure assessment is body burden, not dose. We need to focus on body burden when determining health effects. Jenny Pronczuk: What about looking at microtoxins? These are found in breast milk in African women.

1. Jeff Gearhart: New issues arose in the airforce when women started entering in large numbers. There are four factors to be analyzed:

1. Maternal: Chemical exposure should be analyzed before and during lactation. Other considerations include body adipose levels, milk composition (i.e. fat concentration), age and parity, stored vs. dietary source for milk production.
2. Chemical: Compare lipid solubility of the chemical, ionization (plasma-to-milk gradient), binding of chemicals to maternal blood components, and molecular weight.
3. Physiology: Consider neonate and maternal physiology, in addition to time dependent changes in milk.
4. Breast Milk Physiology: Consider percentage of lipids and protein in the milk as well as the transport functions of the milk.

Time dependent changes in milk: Within feeding sessions, foremilk contains between 1-2% fat, whereas hindmilk contains 4-6% fat. Long term changes include 0-7 days postpartum the milk contains a high protein concentration (10%) and low lipid concentration (1%). Mature milk contains a higher lipid composition (3%) and lower protein composition (1%).

Breast milk contamination with chemicals: Chemicals levels increase with age and decrease with increasing number and duration of lactation periods. Chemical levels in breast milk decrease with increasing number of pregnancies.

PBPK Modeling: This is a tool for deriving estimates of delivered doses of toxic chemicals to the nursing child.

Summary: Milk contamination with organic chemicals is most dependent on the lipid affinity of the chemical and the maternal body clearance of the chemical. Physiologic models allow us to take into account both changes in the maternal physiology and transfer kinetics, as well as the chemical-specific characteristics, in order to produce more accurate estimates of neonatal risk. A mismatch between infant exposure and maternal body burden is possible.

1. Rebecca Clewell: Milk transfer of inorganics is low. Active transport mechanisms can result in high milk:plasma ratios. Synthetic formulas can result in high inorganic chemical exposure (There is Manganese in soy infant formula). Methylmercury is a neurotoxicant found in human food sources. Fish ingestion can result in the accumulation of Mercury in the brain. Perchlorate is an endocrine disrupter found in drinking water. It is involved in confirmed contaminations in 13 states. It has possible neurodevelopmental effects. It also inhibits thyroid uptake of iodine.

Conclusions: Inorganic chemicals may be present in significant amounts (there are active transport systems) in human breast milk. PBPK models enable the use of animal and epidemiological data in human risk assessment.

2. Elaine Faustman: see http://www.nrdc.org/breastmilk for information. This is the National Resources Defense Council’s "Healthy Milk, Healthy Baby: Chemical Pollution and Mother's Milk" web site.

1. Jenny Pronczuk: The goal of this talk is to provide a global perspective on breast milk contamination. The WHO strongly supports breastfeeding. One-third of children under five are malnourished. Malnutrition is linked to 50% of childhood mortality. They have instituted a "baby friendly hospital initiative". See WHA 54.2, the infant and young child resolution. Its goal is to promote six months of breast feeding after birth globally. The WHO is concerned with toxicants and infectious agents in breast milk. Toxic hazards include pharmaceuticals, toxins and toxics, and environmental pollutants. Environmental contaminants include metals, pesticides, and persistent organic toxicants. The WHO has developed the Global Environmental Monitoring System (GEMs). Exposure of infants to organochlorines (DDT) is higher in infants in developing countries where the pesticide is still used. PCB’s exist in higher concentrations in industrialized countries. Risk communication problems are an issue.

1. Phil Landrigan: Spoke about the Ambulatory Pediatric Association (APA), counterpart to the AAP. Discussed the Environmental Pediatrics Fellowship.

Speaking for Deborah Cory-Schlecta: Pesticides have been implicated in Parkinson’s Disease. The synthetic chemical MPTP was shown to cause PD by destroying cells in the substantia nigra region. There is a relationship between Manganese exposure in miners and Parkinsonism. A herbicide called Periclot is chemically similar to MPTP. Environmental exposures can cause neurologic disease later in life once a critical threshold has been reached.

2. Walter Rogan: Described various studies: The Lake Michigan Fish Study, a 1990 study from Holland and Germany, and three contemporary U.S. cohorts. The fetus has no fat to put chemicals into, therefore has less of a body burden. Biomonitoring of breast milk is not a new field. A 1951 study looked at DDT in human milk.

Effects already observed with high toxin concentration in human milk: These infants have a decreased duration of lactation. The infants have tone deficiencies (floppiness). Behavioral differences have been observed in four year olds.

Note: researchers also need to know prenatal exposure when making these determinations. Prenatal effects on IQ are more sever in breast fed kids. Postnatal exposure does not correlate well with prenatal exposure. Exposure measures of endocrine disrupters, such as DDT, are available in some people. Estrogen inhibits the ability of prolactin to promote milk synthesis. Estrogen normally decreases following pregnancy, with prolactin unopposed. Oral contraceptives also decrease the duration of lactation. The strong effect of DDE (a metabolite of DDT) on duration of lactation is something that needs to be confirmed. This may be important in countries with recent spraying of DDT.

Theorized but not observed: direct study of carcinogens not feasible, but better risk assessment may help. No feeding-specific post perinatal mortality data is available in the U.S. It is estimated that 67 days of life are lost from not breastfeeding an infant.

Not yet looked for: Vaccine response. There is reasonable evidence that clinical immunity is preserved. Some evidence (Inuit) exists for increased incidence of otitis media from maternal exposures. A vaccine response is plausible and much easier to quantify than illness exposure.

Resolving confusion: Peripubertal growth and development. PCB and PBB affect girls and boys but not in the same way. DDE makes boys bigger, while phthalates affect breast but no other tissue. When comparing levels with current studies, it is not possible to rank exposure across studies using the published data. Pharmacokinetics: There is a general belief that breast feeding drives toxin levels through adolescence, but not there is not much data.

Question: Is there a way to predict prenatal exposure from postnatal exposure?

There are different effects for different PCB congeners. TEQ’s are not the complete answer. Researchers in Taiwan and Japan have studies the pharmacological use of agents in an attempt to reduce body burdens. Olestra is being investigated as a promising agent for reducing body burdens.

3. Peter Scheidt: In Papua New Guinea, people can only get a bottle of infant formula by prescription.

Overview of National Longitudinal Study: A joint effort of NICHD, CDC, NIEHS, and EPA. Children have an increased vulnerability to environmental exposures. Breastfeeding concentrates fat soluble chemicals. The study was first proposed by the President’s Task Force on Environmental Health in 1998. It was suggested to identify children as early as possible, before birth, and then follow them to adulthood. There are many reasons for this study.

Study Concepts: A high quality national study, with environment defined broadly. It should study a range of common environmental exposures, and less common health outcomes. Also, it should include the environment and genetic expression, state-of-the-art technology, a consortium of multiple agencies, extensive public-private partnerships. The longitudinal study should serve as a national resource for future studies.

Outcomes: Measures include autism-spectrum disorders, ADHD, asthma, hospitalizations, cancer incidences, acute lymphocytic leukemia, CNS tumors, hypospadias, and spina bifida.

The study might be too small for some of these outcomes. The data would be pool the data into NCI cancer registries.

Study Population Issues: It should be generalizable to the U.S. population, it should include additional study populations (women of childbearing age), possible effects of chemicals on fertility and pregnancy. Specific high risk populations include those employed in agriculture, industry, and the economically disadvantaged.

Criteria for Choosing Hypothesese: There should be no single hypothesis. It should consider public health significance, exposure prevalence, required sample size, need for longitudinal study, and measurability. An operating hypothesis is that prenatal/early childhood chemical exposures increase the risk for neurodevelopmental conditions and asthma.

Data Collection: Geographically distributed centers, recruitment, measurement, and follow-up. Advanced technology includes microsampling, the world wide web, geomapping, satellite mapping (for air quality monitoring).

Pilot studies should ensue, the study will commence in Fall 2004 and analyses will be complete by 2030. See http://www.nichd.nih.gov/despr/cohort. Email is nichdcohort@mail.nih.gov.

Question: What about doing this study in developing countries? Answer: He would like to see Mexico involved. The advantages of this include higher levels of lead, DDT, and increased exposures.

1. Kim Hooper: We need a better monitoring system. Breast milk can be used as a matrix for monitoring other exposures. This is a politically powerful tissue. . We need to increase levels of breast feeding in the U.S., while minimizing the public’s exposure to harmful chemicals. There is a need for a study of body burdens in reproductive-aged populations, and community-based monitoring. Hospitals give infant formula to poor people. This is akin to giving cigarettes to teenagers.

WIC: Has a special supplement nutrition program for women, infants, and children. Seven million people participate each month. 62% of infants in California are on WIC. The percentage nationwide is 45%. WIC needs to stop handing out formula for free. The key to increasing breastfeeding is education.

Monitoring systems can warn of excessive exposure, and direct us to initiate a reduction in exposure. The biomonitoring of tissue can serve as a window into the body burden of the community.

Advantages of Biomonitoring of Breast Milk: Noninvasive, easy, convenient, self-instruction, self-collection, looks at persons of reproductive age, encourages breastfeeding, encourages community education.

A California breast milk study in 1998 found ten times the PBBE levels found in Europe. This represents a 100-fold increase over the last ten years. Other studies have examined POP’s in harbor seals. One lesson is that when breast milk "talks", people listen. This is a direct indication of body burden. A good program will establish baseline background levels, hot spots, mitigating factors, health outcome follow-up, time trends, the identification of new chemicals, the establishment of archive samples. A goal is to set up a body burden registry analogous to cancer and birth defect registries. It should look at exposure (not disease) events, should be community based, and include no identifiers. Community data equals community response.

2. George Daston (Discussant): There are two main problems with exposure to persistent bioaccumulative compounds:

1. Trying to reduce the use of such agents. Data on environmental persistence (the driving force) needs to be augmented with human samples. The sample sizes should be increased, as should the number of compounds analyzed.
2. Need new data and methods development: There is a need for models relevant to the human developmental period when breastfeeding occurs. When do the critical exposures occur? We need to do a risk/benefit analysis of breastfeeding under various conditions. We need a quantitative assessment of breast milk vs. formula for risk assessment purposes.

1. Sherry Selevan (Discussant): Risk assessment should lead to risk communication. The challenge is to ask exposure questions and return the results to them without scaring the women participating. We are getting better at measuring more with a little bit, but this will lead us to the point where ubiquitous chemicals chow up in everyone. There are critical windows in relation to body burden and exposure in relation to developmental pathways. See the NRDC web site for a balanced look at breast milk.

Question: Bob Sonawane: What are the three take home points?

Answer: George Daston: We need good hazard data, good exposure information, and the courage of our convictions.

2. Phil Landrigan: Breast milk pushes the limits of the risk assessment paradigm. One problem is the lack of data, coupled with the immediacy of the issue. There is limited data on exposures and long term outcomes. We need to marry risk assessment with the precautionary principle, all the while pushing for more data. The fundamental issue is a lack of data.

Kim Hooper: We should take a lesson from the Swedes. Breastfeeding should be a priority, while limiting the exposure. This should not be done on an individual level.

George Daston: We need the best possible data on hazards and exposure, and risk assessment analysis of the benefits of breastfeeding vs. the risk of exposure to toxic chemicals.

Sherry Selevan: We should add infant formula to the risk assessment analysis.

Landrigan: There is a difference between transient exposure to volatile compounds, and chronic exposure to persistent chemicals.

1. Michael Lerner: The defining issue of the emerging environmental health movement is the right of women to breast feed toxin free. We need a dialogue between all groups, including citizens, industry, and NGO’s. Breast milk is both the best food for infants, and the most toxic food. Breast milk holds a symbolic power unlike anything else. We need to use care in imparting this powerful information. It cannot be suppressed, or offered without great care. Three years ago he needed to negotiate a peace between breastfeeding activists and toxics activists. Through the International POPs Elimination Network, he has worked in many ways to support action.

Take home message:

1. Breast milk is best. Breastfeeding levels should be increased globally.
2. We need to protect the sacred right of women to breastfeed toxin free.
3. Lifestyle change cannot substitute for a global social effort to reduce toxic chemical exposure. We need a national breast milk biomonitoring program. Target levels to reduce chemical levels to should be established. We can identify new toxins, new body burdens of chemicals. We need advocacy, and coalitions of citizens groups, using science as support.

Biodiversity is at its lowest level in 60 million years. Humans are responsible for this "holocaust in life". Climate change, ozone depletion, invasive species, new infectious agents, habitat depletion, and chemical toxics are problems.

Four possibilities: A) Biological chaos B) Achieving sustainability C) Total destruction D) Artificial people on artificial planets

We need a new level of consciousness (like with slavery, human rights, etc.). This step should be made with citizens, government, and industry.

Citizen Group Strategy: The identification of different sectors of chemically affected children. These groups should be linked together with health impacted communities to form a broad-based activist group. This would increase awareness of the need to decrease chemical exposure. Toxins in breast milk is the defining issue in the environmental health movement. Scientists will be looked to for direction. The educational process is going forward. A critical question is who owns science. The three legs; government, industry, and academics all do science. Currently, there is less of a sense that this balance is in place.

A Breast Milk Contradiction: This psychic issue is important. There is a deep linkage between personal and planetary wounds. Analogous to the World Trade Center disaster, the psychic suffering is overwhelming. It is changing the dialogue. We cannot as people to carry additional suffering. We must be responsible about toxins, to not hide from this issue and dialogue.

Abstracts

Session 1. Patterns of Infants’ Exposure to Chemicals in Breast Milk: Analytical Considerations

Judy S. Lakind, Ph.D.

Studies of environmental chemicals in human milk in the United States have provided information which can be used to estimate health benefits and risks to an infant who is breastfed rather than formula fed. These studies have both strengths and weaknesses; assessed individually, they provide snapshots of concentrations of environmental chemicals in the breast milk of (typically) small populations at one time and place. Taken together, it is difficult to make widely applicable statements about levels of environmental chemicals in breast milk in the US because of a lack of consistent sampling methodologies and reporting of the results (LaKind et al., 2001). According to the World Health Organization, the rationale for monitoring breast milk for environmental chemicals is to "…produce more reliable exposure data for risk assessment, to obtain a good overview of exposure levels and trends in different [geographic] areas…, and to identify any specific populations for further follow-up" (WHO, 2000). In order to meet these goals, a consistent protocol must be the basis for future studies of environmental chemicals in breast milk. A carefully planned and executed program of breast milk sampling and analysis would serve to provide the information needed to assess infant exposures during nursing and to provide consistent and scientifically sound information on benefits and risks of breast feeding. A sampling and analysis program on breast milk should include the following goals: obtaining information on women from diverse geographic regions and different socioeconomic backgrounds; analyzing for an increased number of environmental chemicals; obtaining longitudinal information during the course of lactation; making available the results of these studies, with interpretation, in both the peer reviewed literature and more publicly accessible venues; improving the comparability of results across studies, and providing the basis for comparative risk analysis.

An integral part of the planning for a breast milk monitoring program in the US is the development of a protocol describing all aspects of such a program. The task of the participants of the Workshop on Breast Milk Monitoring was to carefully describe the components of a well-conducted breast milk monitoring study, including participant selection, sample collection and analysis techniques, questionnaire development, chemical selection, and data reporting and utilization, specifically for use in the United States. The Panel, comprised of experts in the fields of pediatrics, lactation, breast milk sampling, analytical chemistry, epidemiology, statistics, and comparative risk evaluation, participated in a two-day Workshop during which topics related to the development of a breast milk sampling and analysis protocol for the United States were discussed. The results of the Workshop are being compiled as a published protocol for breast milk monitoring in the U.S.

LaKind, J.S., Berlin, C.M. and Naiman, D.Q. 2001. Infant exposure to chemicals in breast milk in the United States: What we need to learn from a breast milk monitoring program. Environmental Health Perspectives 109:75-88.

WHO (World Health Organization). 2000. Levels of PCBs, PCDDs and PCDFs in human milk. Protocol for third round of exposure studies. World Health Organization European Centre for Environment and Health. March 2000.

Session 1. Patterns of Infants’ Exposure to Chemicals in Breast Milk: Analytical Considerations

Matthew Lorber¹, Linda Phillips²

1National Center for Environmental Assessment, Office of Research and Development, US Environmental Protection Agency, Washington, DC 20460; ² Versar, Inc., 6850 Versar Center, Springfield, VA 22151

The United States Environmental Protection Agency (EPA) estimates that adult exposures to dioxin-like compounds are 1 pg of toxic equivalents (TEQ) per kg body weight per day (1 pg TEQ/kg-day). The dioxin-like compounds include 7 tetrachlorodibenzo-p-dioxin compounds, 10 dibenzofuran compounds, and 13 coplanar PCBs, all of which are considered to have dioxin-like toxicity, and which can be combined into a single concentration with the use of toxicity equivalency factors (TEFs). The 1 pg TEQ/kg-day estimate was derived from an estimate of 65 pg/day exposure to dioxin-like compounds, mainly through consumption of animal food products, combined with a standard assumption of a 70 kg adult. An infant, by contrast, weighs about 3.3 kg and can be exposed to a total of 800 pg TEQ/day by consumption of breast milk, leading to a body weight-based exposure of 242 pg TEQ/kg-day. After one year of breast-feeding, the exposure drops to about 18 pg TEQ/day. This estimates considers declines in the concentrations of dioxins in mothers milk and infant body weight increases.

This paper also evaluates the impact of this exposure to infant body burdens. A simple, one-compartment, first-order pharmacokinetic model is used to predict body burdens. Temporally varying parameters in this model include: infant intakes of TEQs (pg/day), body weight (kg), body lipid fraction (unitless), and the first-order dissipation rate of TEQs in the infant (yr^-1). Constants include the absorption fraction (0.8) and the initial dioxin concentration in infants (10 ppt TEQ lipid-basis, from concentrations measured in stillborn adipose tissue). A rapid first-order elimination rate of TEQ at birth of 1.73 yr^-1 (half-life = 0.4 yrs) is assumed, declining to 0.61 yr^-1 (half-life = 1.1 yr) by one year, and 0.33 yr^-1 (half-life = 2.1 yrs) by 5 years. This TEQ elimination model for infancy and early childhood is based on an approach for TCDD in the literature in which the overall elimination rate is a function of metabolic (enzyme breakdown) and non-metabolic (fecal elimination) processes. At infancy, the high elimination of lipids in the feces leads to a more rapid elimination of dioxin as compared to the more typical TCDD half-life around 7 years measured in adults. As the infant ages to adulthood, the TEQ elimination rate is based on a second literature approach where elimination is a function of body fat fraction; as the body fat increases, the elimination rate slows to a rate of 0.071 yr^-1 (half-life = 9.8 yrs) after 55 years.

Five nursing scenarios are modeled: formula only, 6 weeks nursing, 6 months nursing, one year, and two years. For the 6 and 12 month nursing scenarios, lipid concentrations peak at around 9 weeks at 44 ppt TEQ. The formula-fed infants peak at less than 10 ppt after the first year. This compares to the current adult average body burden of 25 ppt TEQ lipid. Measurements in the literature showed that breast-fed infants body burdens rose above 40 ppt TEQ lipid, while concentrations lower than 10 ppt were found in formula-fed infants. In all four modeled scenarios, the child lipid concentrations merge at about 10 years of age, at a concentration of about 13 ppt TEQ lipid. Key uncertainties and important modeling assumptions include: the rapid elimination of TEQs from the infant, initial conditions in both the infant and mother, and changes in the quality of breast milk as the infant is breast-feeding.

Session 2. Pharmacokinetics of Toxic Chemicals in Breast Milk

Elaine M Faustman, Ph.D.

The focus of this talk is to highlight new advances in molecular and developmental biology that have relevance for improving our understanding of how chemicals can adversely impact development. The talk will build from the recent National Academy of Sciences report entitled "Scientific Frontiers in Developmental Toxicology and Risk Assessment" (http://www.nap.edu/books/0309070864/html) that reviewed these advances and proposed approaches for incorporating such information in our mechanistic investigation of developmental toxicants. A key emphasis of that report to evaluate the significance for toxicology of knowing seventeen key, highly conserved cell signaling pathways essential for development across all species. The talk will also discuss the NRC framework for linking this information into kinetic and dynamic considerations for the evaluation of developmental toxicants. Dr. Faustman will discuss her own research on developmental toxicants using this framework. An important emphasis of the talk will be on how to utilize integrative, multidimensional approaches to improve our understanding of the relevancy of mechanistic research for protecting children’s health.

Session 3. Assessing Outcomes in Children’s Exposures to Toxic Chemicals in Breast Milk

M. Thiruchelvam, E. Richfield and D.A. Cory-Slechta. Departments of Environmental Medicine and Pathology and Laboratory Medicine, University of Rochester Medical School, Rochester, NY.

Pesticide exposure has been considered a risk factor for Parkinsonism (PK), a neurodegenerative disease of motor systems resulting from selective damage to the nigrostriatal dopamine system of the brain. While PK is considered a disorder of aging, it is conceivable that pesticide exposures early in development could adversely impact the nigrostriatal dopamine system, accelerating normal aging processes and/or enhanciing vulnerability to subsequent pesticide exposures. Exposures of adult mice to no effect doses of paraquat + maneb, when combined (PQ+MB), produces selective, potentiated and irreversible neurotoxicity to the nigrostriatal dopamine system, resulting in a model of early environmental PK. To evaluate the hypothesis above, developmental exposures to even lower doses of these chemicals were carried out, with a subset of exposed mice also re-challenged as adults with these chemicals and their sensitivity compared to adult only exposure, and to postnatal only exposures. Developmental exposures to PQ+MB resulted in a progressive reduction in locomotor activity as evaluated at 6 weeks and again at 6 months of age. Mice exposed to PQ+MB developmentally and then re-challenged as adults were even more affected, showing a 70% reduction in motor activity even 2 weeks following the last re-challenge dose. Correspondingly, striatal dopamine (DA) levels were reduced by 37% following developmental exposure to PQ+MB only, but following adult re-challenge DA levels were reduced by 62%. Similar reductions in DOPAC and HVA were also observed. In contrast, no changes in serotonin levels were evident. Developmental exposure to PQ or MB alone produced no significant change in striatal DA or metabolites, but adult re-challenge with PQ or MB significantly decreased levels of DA and metabolites, indicating that silent damage had previously occurred. Total numbers of TH neurons were determined in the substantia nigra, the site of DA cell bodies for the nigrostriatal system 2 weeks after the last adult exposure. Developmental exposure alone reduced TH⁺ cells, with the PQ+MB group showing the greatest decrease. If adult re-challenge also occurred, a further decrease was observed in all the treatment groups, again with the PQ+MB group showing the greatest reduction. These findings indicate that exposures to pesticides alone or in combination early in development can produce permanent and progressive lesions of the nigrostriatal system, and enhance adult susceptibility to environmental pesticide exposure effects. Thus, developmental neurotoxicant exposures may be involved in the induction of neurodegenerative disorders and/or interfere with the normal aging process. Supported by ES10791 & DAMD17-98-1-8628.

Session 3. Assessing Outcomes in Children’s Exposures to Toxic Chemicals in Breast Milk

Rogan WJ, Gladen BC, McKinney JD, Carreras N, Hardy P, Thullen J, Tingelstad J, Tully M. Am J Public Health 1987 Oct;77(10):1294-7

We followed 858 children from birth to one year of age to determine whether the presence of polychlorinated biphenyls (PCBs) and dichlorodiphenyl dichloroethene (DDE) in breast milk affected their growth or health. Neither chemical showed an adverse effect on weight or frequency of physician visits for various illnesses, although differences were seen between breast-fed and bottle-fed children, with bottle-fed children being heavier and having more frequent gastroenteritis and otitis media. Children of mothers with higher levels of DDE were breast-fed for markedly shorter times, but adjustments for possible confounders and biases did not change the findings. In absence of any apparent effect on the health of the children, we speculate that DDE may be interfering with the mother's ability to lactate, possibly because of its estrogenic properties.

Session 4. Research Needs and Implications for Risk Assessment

Kim Hooper, Ph.D.

09 12 01

Monitoring of air or water is a sensible surrogate for human exposure for the great number of organic pollutants that are volatile, water-soluble, or metabolized, and where air and/or water are likely major routes of exposure. Similar surrogates do not exist for the persistent organic pollutants (POPs). POPs have low volatility and water solubility, and air and water are generally minor exposure pathways. These POP chemicals resist environmental or biological degradation, causing them to persist in humans, biota, and the environment. POPs family includes the PCBs, their brominated cousins the PBDEs, and the polychlorinated dioxins, furans, and pesticides.

All life is interconnected via a massive food chain, and POPs biomagnify up this food chain, each trophic level retaining the POPs from the lower level. Humans are at the top of this chain, with infants at the very peak. Lactating mothers utilize their fat stores to produce breast milk, and in so doing mobilize the fat-stored POPs as contaminants into the milk. Daily intake of POPs by a breast-fed infant can be 50-fold higher than adults on a body weight basis.

Some POPs have endocrine-disrupting activity, and cause in humans a wide variety of adverse effects, including cancer, effects on sex ratio, and learning disabilities.

For POPs, humans are a sensible indicator organism, especially for urban contamination. Human milk is a convenient, non-invasive, and simple means of monitoring POP body burdens in reproductive-age populations. Measurement of POP levels in humans is multimedia monitoring, integrating all exposure pathways.

Breast milk monitoring gives an indication of POP body burdens in the community. Geographic variations identify "hot spots" and urban point sources of contamination. Temporal trends can identify new chemicals in need of regulation, and track regulatory effectiveness.

Although monitoring programs exist in several European countries, there is no systematic human milk monitoring in the US or California. Human POP data in California come from research-level monitoring studies. These include:

1. Human milk from Stockton, CA (dioxins, furans, PCBs): 1999
2. Human adipose tissue from San Francisco Bay region (dioxins, furans, PCBs, PBDEs, OCPs): 1999-2001

New classes of POPs are being identified such as PBDEs, pharmaceuticals, musks, and personal care products, and phthalates. Body burden monitoring will characterize the need for regulating and controlling these products.