Skip to Main Content
An official website of the United States government
Epidemiology and Genomics Research Program

Long Island Breast Cancer Study Research Projects

The assessment of environmental exposures and determining their relationship with cancer is difficult. The LIBCSP was a complex research effort that charted new ground in environmental epidemiology. The investigators developed new environmental measurement and laboratory techniques, and explored new ways to study relationships between the environment and breast cancer.

Those who wish to know more about the LIBCSP may be interested in a review paper "Science and Society: The Long Island Breast Cancer Study Project," Nature Reviews Cancer 2005 Dec;5(12):986-94.

Collectively, the LIBCSP consisted of more than 10 studies and the development of a research tool – the Geographic Information System for Breast Cancer Studies on Long Island (LI GIS).

The studies included human population (epidemiologic) research projects, the establishment of a family breast and ovarian cancer registry, and laboratory research on mechanisms of action and susceptibility in development of breast cancer. Findings from all of the studies have been reported.

Some of the projects that comprised the LIBCSP are described below.

Extramural projects:

Breast Cancer and the Environment on Long Island (sometimes referred to as Columbia case-control study)

photo of Dr. Marilie Gammon

Marilie D. Gammon, PhDPrincipal Investigator
Universtiy of North Carolina at Chapel Hill
Chapel Hill, N.C.

The cornerstone of the LIBCSP is the Breast Cancer and the Environment on Long Island Study, which was undertaken to determine whether certain environmental contaminants increase risk of breast cancer among women on Long Island. This investigation has been led by Dr. Marilie Gammon, of the University of North Carolina (UNC) at Chapel Hill, and has been a collaborative effort of New York City and Long Island researchers.

The primary aims were to determine if organochlorine pesticides, including DDT, polychlorinated biphenyls (PCBs), dieldrin, and chlordane, and polycyclic aromatic hydrocarbons (PAH), a ubiquitous pollutant caused by incomplete combustion of various chemicals including diesel fuel and cigarette smoke, are associated with risk for breast cancer among women on Long Island.

Dr. Gammon and colleagues reported the major findings from the study in August 2002. They found that organochlorine compounds are not associated with the elevated rates of breast cancer on Long Island. However, the researchers said that it is possible that breast cancer risk in some individuals may be associated with organochlorine exposures because of individual differences in metabolism and ability to repair DNA damage, and they are continuing to investigate these possibilities. The researchers also found that exposure to PAHs was associated with a modest increased risk for breast cancer.

Following is the UNC press release:

UNC logo

August 6, 2002

Long Island Breast Cancer Study Project uncovers small risk from hydrocarbons, not organochlorines

CHAPEL HILL -- Exposure to air-polluting polycyclic aromatic hydrocarbons (PAHs) in the environment appears to elevate women's risk of breast cancer by a modest 50 percent in Suffolk and Nassau counties, NY, a new much-anticipated study indicates.

The Long Island Breast Cancer Study Project, one of the largest and most comprehensive environmental epidemiologic studies ever done on that cancer, uncovered no increased rate of the illness among area women who might have been exposed to organochlorine compounds.

"Starting with more than 3,000 women in this federally mandated research, we looked at blood samples taken from hundreds of new breast cancer patients and comparable women without breast cancer who served as controls," said principal investigator Dr. Marilie D. Gammon.

Gammon is associate professor of epidemiology at the University of North Carolina at Chapel Hill School of Public Health.

In their work on polycyclic aromatic hydrocarbons, which are known to cause mammary tumors in laboratory rats, researchers focused on PAH-DNA "adducts" - chemicals attached to the genetic material known as DNA, she said. PAHs are inhaled through air pollution, including automobile and airplane exhausts and cigarette smoke, and are ingested by eating selected foods, including grilled and smoked foods. PAH-DNA adducts are evidence of exposure and suspected evidence of tissue damage in humans.

Except for the possible 50 percent increase in the risk of breast cancer associated with the highest levels of hydrocarbon adducts, "there was no increasing elevation in the risk of breast cancer with increasing adduct levels, nor was there a positive association between adduct levels and two of the main sources of PAHs," she said.

"Those are active or passive cigarette smoking and consumption of grilled and smoked foods. Our data indicate that PAH-DNA adduct formation may influence breast cancer development, although the association does not appear to be dose dependent and may have a threshold effect.

"These ambiguous findings shed some doubt on a clear cause-and-effect association," Gammon said. "The study team is continuing to investigate this issue on Long Island through additional research focused on the possibility of individual responses to environmental exposures."

In their research on organochlorines, including the pesticide DDT, its metabolite DDE and industrial compounds known as PCBs, the scientists found no evidence supporting an association between those environmental toxins and heightened risk of breast cancer, she said. Analyzing blood samples from 646 patients and 429 control subjects, they showed such factors as breastfeeding, weight, menopausal status, length of residence on Long Island and form of breast cancer made no difference. Nor was there any increase in breast cancer associated with individual PCB variations, known as congeners. Not excluded was the possibility that such chemicals played an earlier, more subtle role in tumors.

"Recent research by other investigators suggests that organochlroine compounds may be related to the type of breast cancer that has clinical characteristics that are associated with worse survival. This is an important issue that we are continuing to investigate among the women in our study," Gammon said.

A report on the results appears in the August issue of Cancer Epidemiology, Biomarkers & Prevention. Besides Gammon, scientists involved include Drs. Regina M. Santella and Alfred I. Neugut of the Mailman School of Public Health at Columbia University, Mary S. Wolff of Mount Sinai School of Medicine and 24 other scientists from such institutions as Columbia, the State University of New York at Stony Brook, Cornell Medical Center and Long Island hospitals.

"The goal of this population-based, case-control study was to determine whether breast cancer incidence in women in these two counties was associated with exposures to environmental contaminants," said Gammon, deputy director of UNC's Environmental Health and Susceptibility Center and a UNC Lineberger Comprehensive Cancer Center member. "What we observed did not support that possibility strongly."

Because of circumstantial evidence, many experts believe certain pollutants increase women's exposure to the hormone estrogen, which boosts cell turnover so that it's harder for the body to repair damage before cells replicate and pass the damage to new cells, she said. Breast cancer victims are known to have more estrogen in their blood on average than other women.

"We know too, for example, that if a woman's ovaries, which produce estrogen, are removed before she is 35, her risk of breast cancer drops by half," Gammon said. "Men, who have little estrogen, rarely get breast cancer." The increase in breast cancer risk associated with PAH was restricted to women with breast tumors that were either estrogen receptor and progesterone receptor positive or who were negative for both, she said. This means the link was stronger for women with cancers that are considered hormone sensitive. Why is unclear.

"Our findings with polycyclic aromatic hydrocarbons suggest that women's individual responses to similar PAH exposures might be more relevant to breast cancer development than the absolute amount of PAH exposure," Gammon said. "A lot more work needs to be done to sort out exactly what and how environmental exposures may promote breast cancer."

The National Cancer Institute and the National Institute of Environmental Health Sciences paid for the research, which Congress mandated in 1993. The study came in response to concerns by Long Island cancer activists that Nassau and Suffolk counties showed disproportionately high breast cancer levels compared with other parts of New York state.

In a third paper also published in August 2002, Dr. Gammon and colleagues describe the full study population and research methods, and the established risk factors for breast cancer found in the group. Many of the well-known breast cancer risk factors were confirmed in the study. These risk factors included increasing age, having a family history of breast cancer, having a first child at a later age (age 28 or older in this study), never having given birth to a child, and having higher income.


For the population-based study, all women in Nassau and Suffolk counties who were newly diagnosed with breast cancer during a one-year period beginning August 1996 (cases) were invited to participate. A comparison group (controls) of women who did not have breast cancer were randomly selected from the two counties. Altogether about 1,500 cases and 1,500 controls participated. Altogether, 1,508 women who were newly diagnosed with in situ or invasive breast cancer, and a similar number of women who did not have cancer participated in the study. (In situ breast cancer is early cancer that has not spread to neighboring tissue.)

The study participants were asked to answer two questionnaires and to donate blood and urine specimens. The main questionnaire asked the women about their pregnancy and occupational histories, residential history in Nassau and Suffolk counties, use of pesticides in and around the home or farm, use of electrical appliances, lifetime history of eating smoked or grilled foods, active and passive cigarette smoking history, alcohol consumption, medical history, menstrual history, use of hormones, family history of cancer, body size changes by decade of life, lifetime participation in recreational physical activities, and demographic characteristics, such as race/ethnicity. Women were also asked to complete a food questionnaire.

A subset of the study population who had lived in their current homes for 15 years or more was invited to participate in environmental sampling of house dust, tap water, and yard soil. A random sample of white women who had met the residency requirement were invited to participate. All study participants who identified themselves as African American or black and who met the residency requirement were invited to participate. Of the total study population, 383 cases and 429 controls participated in the home study.

Additional Research

Dr. Gammon and colleagues have continued to conduct additional analyses of the data and biospecimens collected for the study and to publish their findings.

In 2001, she received another NCI grant to conduct a follow-up study on 1,098 of the women with breast cancer who had participated in the parent population-based case-control study. This later study investigated whether the environmental factors examined in the parent study and other lifestyle factors influenced disease-free survival and overall survival in Long Island women. Findings from this study also are being published.

Published Reports
Funding / Timeline

Breast Cancer and the Environment on Long Island Follow-up Study

Marilie D. Gammon, PhDPrincipal Investigator
University of North Carolina at Chapel Hill
Chapel Hill, N.C.

Dr. Marilie Gammon, of the University of North Carolina at Chapel Hill, and collaborating scientists in New York City, conducted a follow-up study on 1,098 women with breast cancer who participated in the parent population-based case-control study of Long Island women. The primary aims of the study were to determine whether environmental and other lifestyle factors influence breast cancer survival and overall survival among a population-based sample of Long Island women diagnosed with the disease.

The environmental and lifestyle factors under investigation included organochlorine compounds (including DDT/DDE, PCBs) and polycyclic aromatic hydrocarbons (PAHs), cigarette smoking, physical activity, hormone replacement therapy, adult weight gain, diet, alcohol consumption, and genetic and other personal susceptibility factors. (See parent study for information on DDT/DDE, PCBs, and PAHs.)

Study participants were interviewed and re-asked questions that had been administered as part of the parent study, including about lifestyle factors, personal and family medical history, and occupational and residential history. In this way changes in exposures that had occurred since diagnosis could be examined. Additionally, information about each woman's original breast cancer treatment and any recurrences was collected. The women were followed an average of 5 ½ years after breast cancer diagnosis.

The investigative team continues to analyze data collected in the study and publishes reports on the findings. One finding to emerge from the study indicates that adult weight gain is associated with decreased survival from breast cancer. In this analysis, Dr. Gammon and colleagues found that premenopausal women who gained more than 35 pounds after age 20 and before diagnosis with breast cancer were twice less likely to survive the cancer. Moreover, postmenopausal women who gained more than 29 pounds after age 50 and before diagnosis of the cancer were nearly three times less likely to survive.

"Obesity and weight gain in adulthood are modifiable risk factors for breast cancer occurrence and survival and our results show the importance of weight management, particularly during the perimenopausal and postmenopausal years in the prevention of excess mortality after menopause," write Dr. Gammon and co-authors in the published report of the findings. [Cancer Epidemiology, Biomarkers & Prevention, 2007]

Another finding from the study pertained to PAHs, a pollutant caused by incomplete combustion of various chemicals including diesel fuel and cigarette smoke. In the parent study, the investigative team had found that exposure to PAHs was associated with a modest 50 percent increased risk for breast cancer, which suggested the need for further research in other populations. The scientists had measured the level of binding of PAHs to DNA (forming what is called PAH-DNA adducts) in blood samples taken from the study participants. Formation of PAH-DNA adducts is believed to be necessary for cancer development. In this follow-up study, the investigators conducted a survival analysis that did not find strong support for the possibility of an association between detectable PAH-DNA adducts and an effect on survival among women with breast cancer. [Environmental Research, 2009]

Published Reports
Funding / Timeline

Electromagnetic Fields and Breast Cancer on Long Island

photo of Dr. Leske

M. Cristina Leske, MD, MPH
Stony Brook University, Stony Brook, NY

Dr. Cristina Leske, of Stony Brook University, and colleagues conducted a population-based case-control study to determine if electromagnetic fields (EMF) are associated with increased risk for breast cancer. In 2003, they reported finding no association between residential exposure to EMFs and increased risk for breast cancer.

EMF levels, as measured by in-home spot, 24-hour, ground-current measurements, and wire codes did not differ between women who were diagnosed with breast cancer (cases) and women who did not have the disease (controls). Furthermore, no differences were observed between the two groups when the data were analyzed by controlling for age, family history of breast cancer, personal history of benign (noncancerous) breast disease, number of children (parity), and education.

The study population consisted of 576 women (cases) who were newly diagnosed with breast cancer during the one-year period beginning August 1996, and 585 women (controls) who did not have the disease. All of the women were under 75 years of age and had resided in their current homes in Nassau and Suffolk counties for at least 15 years before the time of diagnosis of breast cancer (cases), or at the time they were identified (controls) for the study. The study population was drawn from the larger study population of the Breast Cancer and the Environment on Long Island Study. (The participation rate was low among women age 75 and older in the larger study, thus they were not included in the EMF study.)

The researchers conducted a comprehensive home assessment of the study participants' exposure to EMFs through personal interviews and by taking a variety of EMF measurements within and around the outside of the home. The 24-hour, spot, and ground-current measurements were taken with specially designed equipment to measure EMFs. EMFs were measured in the following ways:

  • 24-hour measurements were taken in the bedroom and the room the participant spent the most time in, aside from the kitchen (living room, den, most lived-in room);
  • spot measurements were taken at the front door, in the bedroom, and the most lived-in room;
  • ground-current magnetic fields, which are sometimes present in a home's electrical grounding system were measured at the center of the bedroom and the most lived-in room; and
  • wire coding (or wire mapping) was performed to classify the type of wiring surrounding a home and the distance from this wiring; this information was coded using the Wertheimer-Leeper and Kaune-Savitz systems.

These findings are similar to those reported by scientists at Fred Hutchinson Cancer Research Center in 2002. Both of these studies included a comprehensive set of in-home measurement of EMF exposure and wire codes. In addition, Dr. Leske's study included ground-current magnetic field measurements. Furthermore, the study included only long-term residents, with the objective of assessing exposures over an extended time period. The study also is notable for its large sample size and high participation rates, and use of the best available methods to estimate past exposure to magnetic fields in the home.

In additional analyses, the researchers found no association between use of electric blankets and risk of breast cancer. No trends were seen with increased duration of use, frequency of use, or other indicators of more intense exposure to EMFs. Also, electric blanket use was not associated with hormone receptor status of the tumor. The findings from these very comprehensive analyses are consistent with results from most other studies. In an editorial accompanying publication of these results, Dale Sandler, PhD, of NIEHS, commended the study for its high quality.

The researchers also have analyzed data on exposure related to shift work and light-at-night. They found no association between overall shift work and risk for breast cancer. Looking at evening and overnight shift work separately, women who were overnight shift workers were found to be at lower risk for breast cancer than women who never worked shifts. In addition, the researchers found that women who reported rising frequently during the week and turning on lights multiple times per night were at increase risk for the cancer, a finding that has not been reported before.

Plans are to conduct analyses on occupational exposures and use of electrical appliances.

Published Report
Funding / Timeline

Epidemiology of Breast Cancer (including Schoharie County)

photo of Dr. Stellman

Steven D. Stellman, PhD
American Health Foundation, New York, NY

Dr. Steven D. Stellman, of the American Health Foundation, New York, NY, and colleagues conducted a hospital-based case-control study to investigate risk for breast cancer in relation to levels of organochlorine compounds, such as DDT and polychlorinated biphenyls (PCB). For the study, women were recruited who were being treated at North Shore University Hospital, Manhasset; Long Island Jewish Medical Center, New Hyde Park; and Bassett HealthCare, Inc., which serves Schoharie County, an area included in the Congressional mandate for breast cancer research on Long Island. About half of the women who participated in the North Shore University Hospital/Long Island Jewish Medical Center component of the study were residents of Long Island.

Long Island Component

Findings from the North Shore University Hospital and Long Island Jewish Medical Center study component were reported in November 2000. Following is the text of a fact sheet from the American Health Foundation that describes the findings:

American Health Foundation
November 10, 2000

Long Island Study Finds Little Evidence that Organochlorine Compounds Increase Risk for Breast Cancer

Findings from a hospital-based case-control study of women from Long Island suggest that increased risk for breast cancer does not appear to be associated with past exposure to organochlorine compounds, according to Steven D. Stellman, PhD, MPH, of the American Health Foundation (AHF), Valhalla, NY, and colleagues. The lack of association held for both women whose tumors were estrogen receptor positive, as well as for women whose tumors were estrogen receptor negative.

The research findings are consistent with the results of other recent investigations of organochlorine compounds and risk for breast cancer that have been reported in other regions. "No single study is definitive; rather, it is the gathering of evidence from different studies and populations that is necessary to reach a conclusion about whether these compounds are associated with increased risk for breast cancer," said Dr. Stellman. "Other research is under way on Long Island that is examining environmental exposures to organochlorine compounds in relation to risk for breast cancer, and we look forward to the findings from these analyses next year."

The findings are reported in the November issue of Cancer Epidemiology, Biomarkers & Prevention.* Collaborating on the study are: Mirjana V. Djordjevic, PhD, Joshua E. Muscat, PhD, and Lin Gong, PhD, of AHF; Julie A. Britton, PhD, Mount Sinai School of Medicine; Marc L. Citron, MD, and Erna Busch, MD, ProHEaltH Care Associates, Lake Success, NY; and Margaret Kemeny, MD, State University of New York (SUNY) at Stony Brook, Stony Brook, NY Dr. Stellman is also on the Epidemiology faculty at the Mailman School of Public Health, Columbia University, NY.

Study Design

The study included 232 women who had surgery for breast cancer (cases) and 323 women who had surgery for non-cancerous breast disease or for conditions unrelated to the breast (controls). Of the cases, 199 had invasive breast cancer and 33 had carcinoma in situ. The women in the control group had had surgeries involving their gall bladder, removal of lipomas, abdominal hernias, osteoarthritis, and other conditions unrelated to the breast. The women were treated between 1994 and 1996. Both groups were similar in education, race, body mass index (a composite measure of weight and height), age at first live birth of a child, and county of residence.

The women had received their care at Long Island Jewish Medical Center, New Hyde Park, NY, and North Shore University Hospital, Manhasset, NY, which are the two largest hospitals serving the Long Island population. Fifty-seven percent of cases (128 women) and controls (186 women) lived in Nassau or Suffolk County. The other study participants lived in New York City, primarily in Queens County. There were no major differences in levels of the organochlorine compounds and PCBs when the data were compared between women living in Nassau and Suffolk counties with residents of Queens County. Also, no significant differences in exposures were seen between residents of Nassau and Suffolk counties.

The study participants provided adipose (fatty) tissue obtained at surgery, blood samples, and answered a questionnaire about their medical and reproductive history, diet, smoking, and other lifestyle factors. Adipose tissue was used for this analysis because the fat-soluble organochlorine compounds accumulate and are stored in body fat for many years. They are excreted very slowly from the body, making it possible to look for evidence today of exposures that have occurred over a long period. The adipose tissue from the breast cancer patients was obtained prior to chemical or radiation treatment for the disease.

The study focused on 7 organochlorine pesticides and 14 congeners (types) of PCBs. The organochlorine pesticides or their products measured were: DDT and two related chemicals, DDD and DDE (the main breakdown product of DDT in the environment and in the body); oxychlordane and trans-nonachlor, which are products of chlordane, a once-common termite treatment; and two pesticides, β-hexachlorocyclohexane (β-HCH), and hexachlorobenzene (HCB). PCBs are a group of chemical compounds found in coolants and lubricants in transformers, capacitors, other electrical equipment, and some consumer products. DDT was banned from commercial use in the United States in the 1970s, and PCBs are no longer permitted in new equipment; however, these highly persistent compounds are still widely found throughout the environment and in animals and people. Detectable levels of the organochlorine pesticides and PCBs were found in all women studied.


The researchers found that cases and controls had comparable levels of total organochlorine pesticides and total PCBs in their adipose tissue, after adjusting for age and body mass. There was no association between breast cancer risk and levels of total pesticides or total PCBs.

The researchers found an apparent association with increased risk for breast cancer only with the PCB congener 183, which accounts for about 9% of total PCBs. Little is known about this compound's toxicity, except that it weakly induces enzymes which may activate some carcinogens. However, no association was found between risk for breast cancer and the most abundant PCB congener, 153, which is a much stronger inducer and which has also been found to have estrogenic properties. The meaning of the finding for congener 183 is unclear, and the observation needs to be confirmed in other studies now in progress. The research team did not confirm a previously reported association with PCB congener 188.


The researchers continue to follow this study population and have conducted medical follow-ups of the cases to determine whether survival or recurrence of breast cancer may be related to body burden of organochlorines. A questionnaire was mailed to cases in order to obtain data on lifestyle changes and other patient characteristics that may have changed since the women were diagnosed with breast cancer. The women also provided new blood samples to permit study of changes in levels of organochlorine compounds over time.

The research is part of the Long Island Breast Cancer Study Project (LIBCSP), which is a group of studies that are being conducted to investigate environmental contaminants that may be responsible for the elevated rates of breast cancer in Nassau and Suffolk counties (Long Island), Schoharie County, NY, and Tolland County, CT. The LIBCSP is coordinated and funded by the National Cancer Institute (NCI) and National Institute of Environmental Health Sciences (NIEHS).


Beginning in the mid-1970s, findings from human studies began to emerge that suggested that organochlorine compounds or related compounds and PCBs may be associated with increased risk for breast cancer, the researchers explained. More recent studies, however, have not found an association, or only a small suggestion of a possible association, between exposure to the compounds and increased risk for breast cancer. A limitation of studies conducted today though, including the current study, the researchers said, is that a single measurement of body burden of these compounds made at the time of diagnosis may not reflect the cumulative lifetime exposure of individuals or age at exposure, particularly women who may once have had elevated levels of organochlorine compounds which were subsequently eliminated from their bodies. More sophisticated research designs are required which gather comprehensive environmental exposure histories.

The International Agency for Research on Cancer (IARC), which reviews and synthesizes evidence on carcinogenicity of toxic substances and exposures, rates DDT and PCBs as "possible" or "probable" carcinogens, based upon evidence from animal and human studies. Furthermore, some of the organochlorine compounds have been hypothesized to be endocrine disruptors - external agents that interfere with the role of natural hormones in the body, which suggests breast cancer as a potential disease outcome.

# # #

* The study is titled " Breast Cancer Risk in Relation to Adipose Concentrations of Organochlorine Pesticides and Polychlorinated Biphenyls in Long Island, New York." Cancer Epidemiology, Biomarkers & Prevention is published by the American Association for Cancer Research.

Useful Websites:

Schoharie County Component

In the Schoharie County component of the study, 36 women who were diagnosed with invasive or in situ breast cancer between 1995 and 1997, and 56 women who had benign breast disease or other non-breast cancer conditions participated. Blood serum and adipose (fatty) tissue were obtained. This population is too small to permit valid analysis.

Measurement of Body Burden of Organochlorine Compounds

In 1998, Dr. Stellman and colleagues published results of a systematic study of correlations between adipose tissue and serum levels of organochlorine compounds. The analysis validated that either tissue or blood can be used to assess a woman's body burden of organochlorine compounds, a point that has sometimes been challenged. Further, the researchers say, they demonstrated that the "profile" of individual organochlorine compounds found in human tissue was similar to that found in animals, such as fish and birds trapped in wildlife, suggesting similar environmental sources of exposure.

Published Reports
Funding / Timeline

Organochlorine Compounds and Risk of Breast Cancer (including Tolland County)

Tongzhang Zheng, MD, ScD
Yale University, New Haven, CT

Dr. Tongzhang Zheng, of Yale University, New Haven, CT, and colleagues conducted a hospital-based case-control study in Connecticut to investigate risk for breast cancer associated with exposure to organochlorine compounds. Levels of organochlorine compounds are being measured in breast adipose (fatty) tissue and blood serum obtained from women who had surgery or biopsies for breast cancer or benign breast disease.

The organochlorine compounds studied were: hexachlorobenzene (HCB);β-benzene hexachloride (β-BHC); polychlorinated biphenyls (PCBs), a family of chemicals used in coolants and lubricants in transformers, capacitators, and other electrical equipment; and DDT and DDE (a metabolite of DDT), which was once widely used in insect control. Study participants also provided medical and diet histories, and information on alcohol use, cigarette smoking, and other environmental exposures.

The study is considered Long Island Breast Cancer Study Project (LIBCSP)-related because of its potential relevance. As an adjunct to the study, the researchers conducted a pilot case-control study focusing on Tolland County, CT, an area that is included in the Congressional mandate in order to investigate possible environmental causes of breast cancer on Long Island. For this analysis, blood serum and the same types of medical, lifestyle, and risk factor information were collected as for the Connecticut study. The number of women diagnosed with breast cancer was too small to permit valid analysis, however.

The findings from the Connecticut study have been published:

In May 1999, Dr. Zheng and colleagues published separate reports on analyses of two organochlorine compounds. In the first report, which appeared in Cancer Epidemiology, Biomarkers & Prevention, the researchers analyzed risk for breast cancer associated with HCB, and found no increased risk for the cancer associated with the agent. HCB is formed as a by-product in the manufacture of chemicals used as solvents, other chlorine-containing compounds, and pesticides. The study included 304 women who had breast cancer (cases), and 186 women who had benign breast disease (controls) who were treated at Yale-New Haven Hospital between 1994 and 1997 and who were ages 40-79. The researchers did not find a significant difference in HCB levels in the adipose tissue of cases and controls. Further, they did not find significant differences in risk for breast cancer when the data were analyzed according to menopausal status, estrogen or progesterone receptor status, breast cancer histology (cell type), stage of disease at diagnosis, type of benign breast disease, or if the women had breast fed their children.

In the second report, which was published in Cancer, the researchers analyzed the adipose tissue of the same study population for β-BHC, and found no association between risk for the cancer and this chemical compound. They did not find significant differences in levels of β-BHC between cases and controls, nor by menopausal status or, among the breast cancer cases, by estrogen and progesterone receptor status. β-BHC is an isomer (form) of the insecticide benzene hexachloride (BHC), which is also known as hexachlorocyclohexane.

In a third analysis of the same study population, Dr. Zheng and colleagues reported on analysis of the relationship between DDE and DDT exposure and breast cancer risk in the American Journal of Epidemiology. The researchers did not find differences in levels of DDE and DDT in the breast adipose tissue of women who had breast cancer and women who did not have the disease. Thus, these findings do not support an association between adipose levels of DDE and DDT and breast cancer risk.

Published Reports
Funding / Timeline

Reducing Barriers to Use of Breast Cancer Screening

photo of Dr. Lane

Dorothy S. Lane, MD, MPHPrincipal Investigator
Stony Brook University, NY

Dr. Dorothy Lane, of Stony Brook University, investigated whether a telephone counseling intervention aimed at women who are known to underuse breast cancer screening can with, or without, an accompanying educational intervention for their physicians, increase use of breast cancer screening.

She found that women who had a previous mammogram at the start of the study and who subsequently received telephone counseling were more likely to become regular mammography users than women who did not receive counseling. In addition, the findings suggested that the educational intervention for physicians is associated with initiation of mammography use among women who have never had a mammography. More research with larger samples is needed to confirm this finding.

The study found that women who had a previous mammogram at the start of the study and who received telephone counseling were 40 percent more likely than women who did not receive counseling to become regular mammography users by the end of the study. Greater reductions in perceived barriers to mammography were associated with receiving telephone counseling. Telephone counseling did not appear to be effective among women who never had a mammogram. However, data suggested that physicians continuing medical education (CME) intervention has potential for initiating mammography use among women who never had a previous mammogram (described later). Further study with larger samples is needed to confirm this finding.

In a first step of the study, Dr. Lane and colleagues surveyed more than 8,900 Long Island women, ages 50-80, by telephone about their use of breast cancer screening, and over 540 doctors by questionnaire. The researchers identified over 3,400 women who had not received mammograms in the past two years and the past two years prior to that. Forty-five percent of women in the 65 years-of-age and older age group were found to underuse mammography, and 34 percent of women in the 50 to 64 years-of-age group to underuse it.

The chief reasons given for not getting mammograms were procrastination and not believing a mammography was needed — the latter reason given more frequently by women in the older age group. The researchers found that over 80 percent of the women who were underusers of mammography had less than a college degree, the majority (58 percent) had family incomes of under $35,000, and most were homemakers or retired.

During the intervention phase of the study, the effectiveness of telephone counseling and physician education were tested. An annual mailing alerted women that they would be called by the Early Detection Guidelines Education (EDGE) Project and encouraged them to go for mammography. Three annual mailings were conducted during this intervention period and included educational materials and incentives to promote screening.

A three-month period elapsed following each educational mailing before initiating telephone counseling of women who did not return the response sheet, in order to allow them time to obtain a mammogram if they were prompted by the mailings. In the intervention group, 1,165 underuser women reported that 597 mammograms were obtained during the first year of the intervention, and 615 mammograms were obtained during the second year of intervention, for a total of 1,195 mammograms.

A quality control evaluation was also conducted of a sample of 117 counseling calls for the telephone counseling intervention. Seventy-one percent of women who were contacted reported that the calls were "entirely reassuring," and 71 percent of the women said the calls were "entirely helpful." In addition, 65 percent of the women said that they felt "entirely encouraged" by the EDGE counselor to have mammogram in the future. Among women who mentioned a barrier to mammography, 64 percent reported that after discussing it with the EDGE telephone counselor, the barrier would not keep them from getting a mammogram in the future.

Ninety-three physicians obtained the in-office continuing medical education (CME) intervention which included three modules: (1) office systems, (2) behavioral counseling, and (3) revisiting physical examination of the breast. The in-office intervention also included an optional follow-up visit with the office staff and a subsequent visit by a standardized patient.

As an adjunct to the educator office visit, the researchers developed a workbook that can also be used as a freestanding self-instructional continuing medical education (CME) activity. The workbook was mailed to those physicians in the intervention group who did not receive an in-office visit. Physicians' anonymous evaluations of the in-office intervention were very positive.

The researchers found that self-reported overall breast cancer screening educational need scores improved for a greater proportion of physicians in the intervention group, especially among those who received the in-office CME intervention, compared to physicians who did not receive the intervention. Educational need with regard to clinical breast examinations also declined while use of provider reminder systems, preparedness to counsel women about clinical breast exams, and recognition of age as an important risk factor for breast cancer improved in more intervention physicians than control physicians. Findings from the study were reported in1999 to 2002.

Published reports
Funding / Timeline

Metropolitan New York Registry of Breast Cancer Families

photo of Dr. Senie

Ruby T. Senie, PhD, MS
Columbia University, New York, NY

The Metropolitan New York RegistryExternal Web Site Policy was established in 1995 to enroll families with history of breast or ovarian cancer who were willing to participate in research. Both men and women were invited to enroll, with recruitment focused in Long Island (Nassau and Suffolk counties), NY.

Families were eligible to enroll who met one or more of the following criteria: a male relative who had been treated for breast cancer, a female relative who had been treated for breast cancer before the age of 45, a female relative who had been treated for both breast and ovarian cancer, and/or who had two or more first- or second-degree female relatives treated for breast or ovarian cancer at age 45 or older. Participants were asked to contribute personal health information, blood, and urine samples. DNA from members of some families was analyzed to detect mutations (changes) in BRCA1 and BRCA2 genes that may be associated with risk of breast and ovarian cancer. All information was, and continues to be, kept confidential, and all data and specimens are stored separately from any identifying information.

The Metropolitan New York Registry was first directed by Ruby Senie, PhD, of Columbia University (pictured), and funded by the National Cancer Institute (NCI) as part of the Long Island Breast Cancer Study Project (LIBCSP). The Registry and five other NCI-funded Registry sites formed the Breast Cancer Family Registry (B-CFR), which was a major resource of data and biospecimens from high-risk families that is used for many areas of breast and ovarian cancer research, including study of genes and the environment.

This readily available resource of data and biospecimens has helped accelerate research on the causes of breast and ovarian cancer, familial susceptibility, and the impact of environment and lifestyle on cancer development.

In 1999 after five years of funding as part of the LIBCSP, support for the Metropolitan New York Registry was continued as part of NCI's general grant program. Mary Beth Terry, PhD, of Columbia University, presently is the Director. Funding for the six B-CFR sites supports their maintenance as a resource for investigators, follow-up of all participants to collect additional information and biospecimens, and recruitment of additional members of enrolled families.

The Metropolitan New York Registry has enrolled more than 1,400 families including more than 4,800 women and men with a history of breast or ovarian cancer.

All B-CFR sites now are conducting a Follow-up Study of Registry families to collect additional information and biospecimens in order to expand the research resource.

New families are not being recruited, but additional members of families who already are enrolled are invited to join.

To learn more about the Metropolitan New York Registry or the Follow-up Study, call 212-304-6433 or toll free 1-888-METRO-08. The Registry also may be reached by e-mail at:

Funding / Timeline

Environmental Exposures and Breast Cancer on Long Island (within grant titled Environmental and Genetic Determinants of Breast Cancer)

photo of Dr. O'Leary

Erin O'Leary, PhD
Stony Brook University (funded through the grant "Environmental and Genetic Determinants of Breast Cancer")

Dr. Erin O'Leary, while at the University of Buffalo, conducted a nested, case-control study to determine if residence in close proximity to hazardous waste sites, toxic release inventory sites, prior land use (for example, farm land), and exposure to various chemicals in drinking water may be associated with breast cancer on Long Island.

The study population was selected from a cohort of New York State residents in 1980, established by investigators at the University at Buffalo, who had lived at least 18 years in their current residences and had completed a mailed questionnaire. Within the cohort, 3,097 women from Long Island answered the questionnaire. From this Long Island group, data on 105 women who had been diagnosed with breast cancer (cases) between 1980 and 1992 were compared to data on 210 randomly selected Long Island women who did not have breast cancer (controls), and who were age and race matched to cases.

Because there is no proven way to measure an individual's historical environmental exposure to most chemicals in the environment, Dr. O'Leary relied on proxies to estimate exposure. She linked data on each woman's residence with data on historical land use, drinking water, and proximity to hazardous waste sites and toxic release inventory sites, and estimated historical exposures to organochlorine and carbamate pesticides, solvents or volatile organic compounds (VOCs) (for example, tetrachloroethylene, trichloroethylene, and 1,1,1-trichloroethane), nitrates, and metals (cadmium, chromium, and arsenic). Geographic information software was used to calculate the distance between residences and the point sources of pollution.

Findings from this exploratory study are:

  • Women who lived within 1 mile of hazardous waste sites containing organochlorine pesticides were found to have an increased risk for breast cancer, after adjusting for known risk factors. No association was found between breast cancer and pesticides detected in drinking water or residence on or near agricultural land. Women who lived on or near agricultural land and who were nulliparous (never given birth) or had an older age at first birth had an increased breast cancer risk, compared to women who did not live on or near agricultural land and who had a younger age at first birth. The small sample size limited the study power, and the confidence intervals were wide, which means the findings could be due to chance. These findings suggest the need for additional research on the topic.
  • No association was found between increased risk for breast cancer and exposure to VOCs in drinking water, or residing in close proximity to hazardous waste sites and toxic release inventory sites containing these compounds. A recent hypothesis suggests that VOCs may be transformed in the breast fat tissue, excreted into the ductular systems, and may initiate or promote breast carcinogenesis.
  • No association was found between increased risk for breast cancer and levels of nitrates or metals in drinking water. None of the wells in the water districts studied exceeded the maximum contaminant levels for these compounds allowed by the U.S. Environmental Protection Agency. Also, risk for breast cancer was not associated with living within 1 mile of hazardous waste or toxic release inventory sites containing metals. (Nitrates were studied because they have been linked to other cancers, but not specifically to breast cancer. One occupational study indicated a slightly increased risk for breast cancer in women exposed to metals. In another study, the metal cadmium was found to increase the growth of human breast cancer lines in cell culture and may have estrogenic activity. Results of a recent study show that cadmium has estrogen-like activity in rats (in vivo).)

Published Report
Funding / Timeline

Estrogen Metabolites as Biomarkers for Breast Cancer Risk (also called Breast Cancer Risk and Inducibility of P450s)

photo of Dr. Bradlow

(Also known as Breast Cancer Risk and Inducibility of P450s)
H. Leon Bradlow, PhD, Principal Investigator
Strang Cancer Prevention Laboratory, New York, NY

Long Island investigators investigated whether differences in the way women's bodies process the natural hormone estrogen may be related to breast cancer risk. Estrogen is metabolized by two main, competing pathways, either to 2-hydroxyestrone or to 16α-hydroxyestrone. Earlier studies suggested that the balance between the estrogen metabolite 16α-hydroxyestrone, which has been associated with breast cancer, and 2-hydroxyestrone, which has not, may affect risk for the disease. Dr. H. Leon Bradlow, of Strang Cancer Research Laboratory, and colleagues examined the ratio between the two metabolites in urine. They found that postmenopausal women with very low levels of the "good" metabolite relative to the "bad" metabolite had a greatly increased risk of breast cancer, compared to women with high levels of the "good" metabolite. Because of the small number of study participants, further study is needed to confirm the findings.

Published Reports
Funding / Timeline

Regulation of Scatter Factor Expression in Breast Cancer

Eliot M. Rosen, MD, PhD
Long Island Jewish Medical Center,
Albert Einstein College of Medicine, New Hyde Park, NY

Dr. Eliot Rosen and colleagues, of Albert Einstein College of Medicine, New Hyde Park, NY, evaluated how scatter factor, a growth factor, may regulate the growth of human breast cancers. Greater knowledge of the mechanisms important for breast cancer development and progression is of keen scientific interest. Such information can provide insights to better understand the nature of the disease and to develop new therapies to prevent or halt its progression.

Most human breast cancer cells contain high levels of the receptor for scatter factor, which means that, in the laboratory, breast cancer cells will vigorously multiply in the presence of the growth factor. Dr. Rosen and his colleagues found that scatter factor causes human breast cancer cells to move faster and to be more invasive in cell cultures. The growth factor induces the breast cells to produce an enzyme that degrades tissue, they found, thus facilitating tumor invasion.

In laboratory animals, they found that scatter factor greatly stimulates the formation of new blood vessels (angiogenesis), an essential step for tumor growth and spread (metastasize). They also found that invasive breast cancer tissue samples with higher levels of scatter factor have higher levels of von Willebrand's factor (VWF), which is a protein produced by the lining of blood vessels. This suggests that higher levels of VWF may be associated with greater angiogenesis.

In other experiments, the investigators found that levels of scatter factor were significantly higher in invasive breast cancer tissue than in benign breast lesions or non-invasive breast cancers (ductal carcinoma in situ (DCIS)). Invasive breast cancers had nearly four times the scatter factor content of DCIS tissue. And tumors that had spread to the axillary lymph nodes had higher levels of scatter factor than invasive cancers that had not yet spread beyond the breast, although this difference was not as great as that seen between invasive breast cancers and DCIS tissue. Scatter factor levels did not vary by histologic (cell) type of invasive breast cancer.

Dr. Rosen and his colleagues also demonstrated in an animal model that scatter factor can stimulate the growth of human breast tumors. Human breast cancer cells altered so that they produced high levels of scatter factor were injected into mice. The researchers found that mammary (breast) tumors grew much more rapidly in these mice than in those that received unaltered breast cancer cells.

Tissue specimens from the Long Island Jewish Frozen Tumor Bank at Long Island Jewish Medical Center were used for the research.

Findings were reported in 1996. Since this study, Dr. Rosen has continued his research on scatter factor further advancing understanding of its role in tumor development.

Published report
Funding / Timeline

RDA Analysis of Breast Cancer

Michael H. Wigler, PhD
Cold Spring Harbor Laboratory
Cold Spring Harbor, NY

Dr. Michael Wigler, of Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, has been examining genetic changes in breast tumor tissue from patients on Long Island using a technique called representational difference analysis (RDA). In this pilot study, he demonstrated that the technique permits cloning of DNA probes that mark the presence of genetic lesions in tumors. Advances in the development of this technique as a research tool are of interest because, among other reasons, because certain point mutations in genes may be linked to environmental exposures and may be characteristic of specific environmental exposures. A report on the pilot work funded through by the Long Island Breast Cancer Study Project (LIBCSP) was published in 1995. The seed money provided for this study led to a larger research effort that was funded by the U.S. Department of Defense.

Published report
Funding / Timeline

Intramural projects:

New Statistical Methodology for Determining Cancer Clusters

Martin Kulldorff, PhD
University of Connecticut Health Center, School of Medicine
Farmington, CT

Dr. Martin Kulldorff, while at the National Cancer Institute (NCI), Bethesda, Md., and colleagues developed an innovative statistical technique that shows that women living in a broad stretch of the metropolitan northeastern United States, which includes Long Island, are slightly more likely to die from breast cancer than women in other parts of the Northeast. The 1997 study does not explain why these women are at higher risk of death, and the researchers note that the increase may be due to differences in well-established risk factors for breast cancer which they were unable to include in the analysis. The researchers found that the breast cancer mortality rate along a section of the East Coast stretching from New York City to Philadelphia was 7.4 percent higher than the rest of the Northeast. They urged caution in interpreting studies of geographic clusters in cancer mortality. The study is described further in an NCI backgrounder. Dr. Kulldorff is now at the University of Connecticut Health Center.

Published Report

Geographic Influences on Women's Health

Susan R. Sturgeon, DrPH
University of Massachusetts
School of Public Health, Amherst, Mass.

Dr. Susan Sturgeon, of the University of Massachusetts School of Public Health, Amherst, Mass., conducted a study to determine the extent to which variation in the prevalence of breast cancer risk factors was responsible for regional differences in breast cancer mortality rates. She began this study while working at the National Cancer Institute.

The study focused chiefly on geographic variations among white women in the Northeast and South — where differences in breast cancer death rates are most pronounced, and on rural and urban variations in rates in these regions.

Breast cancer death rates for white women vary substantially by region of the country and population density. In 1985-1989, rates for women in the Northeast were 20 percent higher than in the South, which had the lowest rates. Furthermore, rates tended to be higher in urban than rural areas within the same regions of the country. They also varied within counties with similar population densities and within the same region. Breast cancer death rates for black women did not differ as markedly by geographic area as for white women.

Dr. Sturgeon's study included 3,800 women, 25 years-of-age or older (selected by random digit dialing). About 300 women were from Nassau County, NY Between December 1994 and April 1995, the research team interviewed study participants by telephone for information about their age, menstrual and reproductive history, personal and family history of breast cancer, history of benign breast biopsies, exogenous estrogen use, body mass index, dietary fat intake, and alcohol consumption. The potential contribution of differences in the prevalence of breast cancer risk factors to population-density and count-level variation in the 1990-1994 breast cancer mortality rates was then analyzed.

Publishing their findings in 2003, the researchers reported that "higher breast cancer mortality rates in high-density areas relative to low-density areas among women 55 years and older in the Northeast and South were completely explained by the higher proportion of women with established breast cancer risk and prognostic factors in high-density areas. The most important factor was age at first livebirth." Among younger women, they added, little or no variation in breast cancer mortality rates were observed between low- and high-density areas either before or after adjustment for established breast cancer risk factors.

In addition, the researchers said, "County-level variation in breast cancer mortality rates in 1970-1979 within high-and low-density areas of the Northeast and South was considerably attenuated [weakened] when 1990-1994 mortality rates from the same group of counties were used. The same attenuation in the mortality ratios over time was seen for Nassau County. There was no statistically significant excess mortality in Nassau County compared to low-mortality, high-density areas in the Northeast using 1990-1994 rates after adjustment for known risk factors."

The Northeast area covered in the study includes Connecticut, Delaware, Maine, Maryland, Massachusetts, New Jersey, New Hampshire, New York, Pennsylvania, Rhode Island, Vermont, and the District of Columbia. The southern states are Alabama, Arkansas, Georgia, Kentucky, Louisiana, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, and West Virginia.


The study above was a follow-up to a 1995 analysis by Dr. Sturgeon and colleagues of geographic differences in breast cancer death rates that demonstrated that the mortality risk for breast cancer among white women is similar in the Northeast, Midwest, and West, after recognized risk factors for the disease are taken into account. Moreover, the researchers found that the differences between these regions and the South narrowed after accounting for such factors. The study did not explain all the regional differences in breast cancer, but provided perspective and laid the groundwork for the present study.

The findings appeared in "Geographic Variation in Mortality from Breast Cancer Among White Women in the United States," Journal of the National Cancer Institute, by Drs. Sturgeon, Catherine Schairer, Mitchell Gail, et. al, December 20, 1995, vol. 87, no. 24, p. 1846-1853, along with an editorial, "Geographic Patterns of Breast Cancer Among Women," by Drs. William Blot and Joseph McLaughlin, p. 1819-1820 (no abstract available). Press Release.

Published Report

New York State Cancer Registration Project

Brenda Edwards, PhD
National Cancer Institute, Bethesda, Md.

Dr. Brenda Edwards, of the National Cancer Institute (NCI), worked with the New York State Department of Health and Long Island cancer registrars to review state cancer registration and reporting procedures for breast cancer. She conducted a study of medical records of breast cancer cases on Long Island, and looked at the accuracy of reported stage of diagnosis for female residents using the New York Cancer Registry and a report from the Centers for Disease Control and Prevention (CDC) on data from 1978-1982, which indicated a high percentage (18 percent) of distant disease for Long Island. A 10 percent oversample was used to obtain 100 cases per year per county to reach a total of 600 cases. For 1989-1991 in Long Island, the stage distribution showed a smaller percentage of metastatic disease (7 percent) than in the older reported data (18 percent for 1978-1982). When comparing data reported to New York state to U.S. data, overall agreement was good (73 percent). However, misclassification was found at all stages and in both directions (i.e., upstaging and downstaging). The final adjusted stage distribution for 1989 and 1991 showed there was somewhat more regional disease in Long Island than was found in national data among white women.