Summary of the Third Meeting of the Ad Hoc Advisory Committee
Wednesday, September 24
Environmental Exposures and Breast Cancer on Long Island
Ms. Erin O'Leary, now of the University Medical Center at Stony Brook, explained that her research focuses on past environmental exposures to pesticides or chemicals that are potentially linked to long-term residency near farm land or point sources of industrial contamination on Long Island, or in a water district that has experienced inorganic or organic contamination, including pesticides. The case-control study is being conducted as her dissertation research project, under Dr. Jo Freudenheim, of the State University of New York at Buffalo. Data are being compared on 105 women in Nassau and Suffolk Counties who developed breast cancer between 1980 and 1992, with data on twice as many age-matched controls. No direct contact with the study participants is required. All study participants were among 3,097 women who participated in a 1980 survey of long-term residents. Most of the Long Island women with breast cancer (95 percent) were diagnosed when they were more than 50 years of age. Seventy percent of cases and controls had lived in their homes (prior to 1980) for greater than 20 years, and 96 percent had lived in their homes at least 10 years. Ms. O'Leary anticipated completing the data gathering, analysis, and thesis writing by mid-1998.
No public records of pesticide use were available, so residence on or near farm land was used as a surrogate, and aerial photographs from 1947-1957 were used to distinguish land uses. For point sources of industrial contamination, Ms. O'Leary used a file provided by the New York State Department of Health, which included the New York State Industrial Directories, to identify businesses with 20 or more workers in chemical, metal, electrical, transportation, machinery, equipment manufacturing, and scientific and professional equipment industries. For the hazardous waste sites, which were found on a list of inactive hazardous waste disposal sites in the two counties, two state agencies provided information on specific chemicals and dates of usage of each site. The earliest data available on toxic release inventories is from 1988. Ms. O'Leary assumes that a greater amount of chemicals was released by these sites in the 1960s and 1970s. Using geographic information software, calculations are made of the distances of residences from the various chemical sources.
Ms. O'Leary also is charting the location of residences with respect to Nassau County water districts and Suffolk County well fields. Usually the water supply to a particular home comes from a mixture of water pumped from several wells. She looks at the water supply records to identify chemicals reported to exceed the exposure guidelines. The most common organic chemicals to exceed guidelines are volatile organics, such as tetrachloroethylene and trichloroethylene. These have been hypothesized to cause mutations that can result in cancer. Among the pesticides and other organdies that have been detected in wells are heptachlor, heptachloroepoxide, 2,4-D, dieldrin, chlordane, lindane, 1,2-dichloropropane, and aldicarb. Inorganic chemicals (nitrates, cadmium, arsenic, and chromium) are being looked into because they have been associated with estrogenic, mutagenic, or carcinogenic activity in animals and/or humans.
Discussion: Ms. O'Leary said that the 1980 questionnaire was mailed by researchers from the State University of New York at Buffalo to 265,000 New York State residents identified by address or phone number as living at the same residence for 18 years or more prior to 1980. The brief questionnaire focused on diet and cancer.
Meeting participants asked about home use of pesticides and using oral history as a source of information on home exposures and potential nearby point sources of chemicals. If Ms. O'Leary should do additional research, Dr. Allen offered to speak with her about some pilot work reconstructing the patterns of use of pesticides from a combination of surveys and the uses prescribed on the label.
Responding to concern about the low number of African American women in the study (3 of the 105 cases), Ms. O'Leary said that selection had been based strictly on length of residence. The percentage of African Americans in the original survey was 3 percent.
A question was raised about whether her results could become a component of the GIS for the LIBCSP.
Cape Cod Breast Cancer and Environment Study
Dr. Julia Brody, The Silent Spring Institute, said that Cape Cod and Long Island share certain geologic features, such as sand dunes and glacial deposits. The incidence rate for breast cancer on Cape Cod is actually higher—for example, Barnstable County's annual rate was 136 new cases per 100,000 population at a time when the rates in Nassau and Suffolk Counties were 117 and 111, respectively. The Cape Cod Study was initiated when activists called the state legislature's attention to the elevated breast cancer incidence in the area, which, for most towns, is at least 20 percent higher than for the rest of the state. The legislature then funded a $3.6 million, 3-year study. The Silent Spring Institute, which has a multidisciplinary staff, serves as the study's team leader, and is advised by Public and Scientific Advisory Committees. There is also an independent, state-appointed Peer Review Committee.
The first two goals for the study were to describe the patterns of breast cancer on Cape Cod, and to investigate distinguishing environmental features that might explain its incidence. One feature noticed early during data collection was that the towns with a higher incidence of the cancer tend to be located on the outer part of the Cape. Consequently, the Superfund site on the upper Cape (nearer to the mainland of the state) could not be considered responsible. Study staff also eliminated stage of diagnosis by mammography as a determining difference. A logistic regression analysis using case-control data from The Collaborative Breast Cancer Study indicated that the Cape continues to have a higher incidence rate when traditional risk factors, such as family history of breast cancer, menstrual history, childbearing, smoking, alcohol use, and lactation have been taken into account.
To look at possible environmental factors that may be responsible for the elevated rates, the study team decided to focus on pesticide use and waste water. Although the Cape has a history of widespread agricultural pesticide use, its heaviest use of pesticides was in the forests for eradicating gypsy moths. The interest in waste water relates to the Cape's hydrogeology—a shallow aquifer under sandy soil. Septic tanks are used widely, and waste water is disposed on land to protect the marine sanctuary along the coastline. The potential exists, therefore, for migration of pollutants into the aquifer, which is the drinking water source.
The study's GIS dates back to 1951 for most of its data sources. The land use maps, which are based on aerial surveys, have been particularly useful. The GIS contains spreadsheets linking detailed data to each measurement shown on the maps. Information is included on land use, public water supplies, well locations, well pumping, groundwater contours, water distribution systems, extent of vinyl-lined asbestos cement in piping, nitrates, volatile organic compounds, and pesticide use as well as some information on the 13,000 private wells on the Cape. Tax records are used to map addresses. Using overhead transparencies, Dr. Brody displayed some maps demonstrating the utility of layering data. It is possible to assess individual exposures with the GIS information, she said.
The Cape Cod Study also conducted a limited field sampling program of waste water, ground water, and drinking water for about 225 chemicals—including estrogenic compounds, nitrates, other inorganic compounds, and volatile and some semi-volatile organic compounds, and compared the results to historical records. As expected, alkylphenols and estrogenic activity were found in sewage. Many of these compounds moved into ground water from the waste water. Very low concentrations of some estrogenic target compounds were identified in a few of the 27 private drinking water wells tested.
The study team is currently exploring statistical techniques
for combining the environmental and health data. Dr. Brody invited
meeting participants to visit the project offices to see the database
being applied to answer questions. The Silent Spring Institute's
Web site address is: http://www.silentspring.org
.
Dr. Brody also provided copies of "Mapping Out a Search for Environmental
Causes of Breast Cancer," by Dr. Brody et al., in Public Health
Reports, vol. 3, no. 6, Nov./Dec. 1996.
Discussion: Asked about contaminated surface water, Dr. Brody said that the nearby ocean has been a marine sanctuary for many years, but there is some coastal bay contamination. Learning the fate of the contaminants as they move through the environment is an emerging area of study.
Dr. Brody explained that the land area contributing water to a well is called the zone of contribution, and mapping these areas shows how land use may affect water quality. Comparing the drinking water on the Cape to that of other areas of Massachusetts has not been useful, because most other areas are served by surface water. For on-and-off Cape comparisons, the study team would like to have additional information about historical pesticide use in the rest of the state, but off-Cape pesticide use records have not been systematically reviewed.
Asked about confidentiality policies when it appears possible to apply the GIS at the level of individual households, Dr. Brody explained that access to the GIS is limited to the team and associated researchers who understand the data sources and limitations. Point maps used in public never show data at a level below the 5-person minimum standard set by the state cancer registry. She noted that most of the data are already publicly available through other sources, but without the composite assembly that the GIS provides.
The Public Advisory Committee has played an important role in the development of the GIS, Dr. Brody said, including in decision making on the kinds of data to be incorporated. This decision making is an on-going process as new research questions are raised.
Quality control and quality assurance procedures are used to verify the data, depending on the source. Data have been received promptly from some sources, and only after enormous lags from others.
Someone said it would be helpful to be able to validate historical exposure data with current measurements taken in homes for which exposures are being estimated. Dr. Brody replied that they would like to have the financial resources to be able to do this. There was also some discussion about need for a more sophisticated model of exposure to pesticides from cranberry bogs. A previous study showed an association between brain cancer and distance from the bog, but did not find an association for breast cancer. Some of the heaviest use of pesticides on the Cape was in cranberry bogs.
Responding to a query, Dr. Brody said that the Pilgrim nuclear power plant is an unlikely major factor in causation of the elevated breast cancer rate because the plant is located just off the Cape, and not near the outer Cape where the incidence rate for the cancer is the highest. Also, the prevailing winds at the Pilgrim site blow away from the Cape. Breast cancer rates are not elevated closer to Pilgrim and in towns downwind.
