Skip to main content
         

Vol. 10 Issue 2, Spring 2005

Pesticide Use and Breast Cancer Risk among Farmers' Wives in the Agricultural Health Study
The Ribbon 

Suzanne Snedeker, Ph.D.
Associate Director for Translational Research, BCERF
Sprecher Institute for Comparative Cancer Research

Engel, Lawrence S., Hill, D.A., Hoppin, J.A., Lubin, J.H., Lynch, C.F., Pierce, J., Samanic, C., Sandler, D.P., Blair, A., Alavanja, M.C. Pesticide Use and Breast Cancer Risk among Farmers' Wives in the Agricultural Health Study. Am J Epidemiol 161: 121-135, 2005.

Most studies that have evaluated whether pesticide exposure from farming activities affects cancer risk have been conducted in male farmers. Because male farmers can have a higher incidence of lymphomas, leukemias, multiple myeloma, and stomach, brain, and prostate cancer, there has been great interest in deciphering whether these cancers are linked to exposures to specific pesticides, and if pesticide exposures may be linked to cancers in farmwomen. The few cancer studies conducted on women who live or work on a farm have yielded limited information because only small groups of women were included, and most studies did not collect in depth information on exposures and corresponding health effects to different types of pesticides or to individual pesticides. This large scale, long-term study of Iowa and North Carolina farmwomen is the most comprehensive effort to date devoted to addressing the complex question of whether specific pesticides do or do not increase the risk of breast cancer. This study, part of the Agricultural Health Study (AHS), overviewed in this issue of The Ribbon, is unique on many levels.

Unique Aspects of this Study

First, the study followed over 30,454 farmer's wives in two states over a long period of time. Women with no history of breast cancer were enrolled during 1993-1997, and information was gathered by questionnaire on use of pesticides by the women and by their husbands. By the year 2000, 309 women were diagnosed with breast cancer. During this time, only 0.2% of the women originally enrolled moved out of their state.

Secondly, information was gathered on the use of 50 different pesticides. This is without precedent. While studies in the past have focused on organochlorine pesticides and breast cancer risk, this study evaluated both contemporary and phased out persistent pesticides. Investigators were able to evaluate the risk of each pesticide and different types of pesticides (herbicides, insecticides, fungicides and fumigants), as well as ten different chemical classes. Some of the chemical classes and examples of pesticides include organochlorines (DDT, chlordane and dieldrin), phenoxyherbicides (2,4-D, 2,4,5-T, 2,4,5-TP], triazines (atrazine and cyanazine), organophosphates (dichlorvos and chlorpyrifos) and many other pesticides for which we have had no or little information.

Thirdly, detailed information was obtained on how long and often the agricultural pesticides were used, how long the family had lived on the farm, how far their house was from where the pesticides were used, and other exposures such as laundering of work clothing, and household use of pesticides. Information was also obtained on other factors that can affect breast cancer risk, such as reproductive history, whether the women was pre- or post menopausal, parity (number of children), family history of breast cancer, as well as body size, physical activity, diet, smoking, alcohol use, and other lifestyle factors. Breast cancer risk was evaluated with respect to both to the wife's use of pesticides, and to the husband's pesticide use in the wives who never used pesticides.

Summary of Results by Pesticide Type

The results of this large-scale study of breast cancer risk in farmer's wives from Iowa and North Carolina suggests the use of the majority of 50 pesticides studied did not increase breast cancer risk. However, use of some specific pesticides did show a small to moderate increase in the wives' breast cancer risk.

Phenoxy herbicides. The most consistent evidence was for a phenoxy herbicide no longer used called 2,4,5-TP. Risk was increased 1.9 fold in women from Iowa, and 2.1 fold in North Carolina women whose husbands used this pesticide. This pesticide was known to be contaminated with dioxin. There is some evidence from an industrial accident of dioxin exposure in Italian women that exposure to this environmental contaminant early in life may lead to a higher breast cancer risk later in life. How and to what extent dioxin may influence breast cancer risk at critical life stages is an area of ongoing research. It should be noted that in the AHS, breast cancer risk was not increased through the farmer wives' use or husbands' use of the phenoxyherbicide in current use called 2,4-D. This result suggests it is more likely that the increased breast cancer risk associated with the use of 2,4,5-TP may be due to dioxin contamination of this herbicide, than to the herbicide itself. The 2,4-D in current use is not contaminated with dioxin.

Organochlorine insecticide and fungicide. The use of the persistent organochlorine insecticide dieldrin and the fungicide captan were also associated with a 2.0 to 2.7 fold (respectively) increase in breast cancer risk in the women enrolled in this study. This was observed only in the farmwomen whose husbands used these pesticides. Use of dieldrin or captan by the women themselves did not affect their breast cancer risk.

One criteria for evaluating cancer risk is whether incidence of the disease increases as exposure increases. Dieldrin was one of the pesticides identified in this study as having higher breast cancer risk with increasing cumulative years of use. To date, only a few studies have evaluated whether dieldrin exposure is related to breast cancer risk in humans, with some studies showing an association (Høyer et al. 1998), while others have not. This study suggests further research is needed to clarify whether this very persistent organochlorine insecticide affects breast cancer risk.

Captan has been classified by EPA as a "probable carcinogen" because it is known to induce tumors in the digestive track in mice, and to a lesser extent induce kidney and uterine tumors in rats. It has not been associated with causing mammary (breast) tumors in animal cancer bioassays. This is the first report of a positive association of captan use with human breast cancer. While captan can cause damage to DNA (mutagen), the biological mechanism for increasing breast cancer risk is not known.

Other organochlorine insecticides. While there was some evidence of increased breast cancer risk for several organochlorine insecticides (aldrin, carbaryl, chlordane, heptachlor, lindane and malathion), the risks were low, not consistent from state to state, and need further conformation before firm conclusions can be made.

Use of many other pesticides, including DDT, and high-use pesticides such as atrazine, alachlor, 2,4-D, and glyphosate, failed to show a significantly higher risk breast cancer risk in farmwomen.

Organophosphate insecticides. Three organophosphate insecticides (chlorpyrifos, dichlorvos, and terbufos) were associated with a higher risk of breast cancer only in premenopausal women, and only in women who used these pesticides. Use of these pesticides by their husbands had no effect on their spouses' risk of breast cancer (relative risk: chlorpyrifos 2.2; dichlorvos 2.3; terbufos 2.6). In past studies, there has not been consistent evidence of environmental chemicals increasing breast cancer risk in premenopausal versus postmenopausal women. While there is some evidence that dichlorvos induces a type of mammary tumor in rodents, in general few studies have been done to evaluate whether this chemical class of commonly used insecticides affects breast cancer risk in women. BCERF researchers recommended the need for further investigations of organophosphates' potential to affect breast cancer risk in Critical Evaluations published in Comments in Toxicology several years ago, and we are encouraged that investigations of pesticides and breast cancer risk are now expanding to include organophosphates. Organophosphates have been reviewed extensively by the EPA. Chlorpyrifos, a known neurotoxin, had been used for many years in homes for termite control. While this use is being phased out, its use to protect crops is still allowed, with continued potential exposure to farm families.

A higher breast cancer risk for the women's use of diazinon (relative risk 1.7), and husbands' use of parathion (relative risk 4.2) and parquat (relative risk 3.9) was only observed in a very small group of women with a family history of breast cancer. This suggests more work needs to be done to confirm this preliminary finding, and to see if there is a gene-environment interaction that may explain these links between family history, breast cancer risk and use of these pesticides.

Further Considerations

While this study stands as one of the few sources of information on the breast cancer risk of a wide assortment of pesticides used currently and in the past, it does have limitations. This study did not report on actual exposure to the pesticides, but used questionnaires to estimate past use in wives and husbands. There may have been problems in accurately recalling past use, resulting in either under or over estimating actual lifetime usage patterns.

The majority of the pesticides that were identified as having a higher breast cancer risk in wives were linked to use of the pesticides by their husbands. The only pesticides identified as being used by the farmwomen themselves that were associated with a higher breast cancer risk were three organophosphates used by premenopausal women. There were no other instances where pesticides used by the farmwomen resulted in a higher breast cancer risk. Why was there an inconsistency? Is this because actual exposures were lower in women because they tended to use protective equipment more than husbands, or that their frequency of use was lower compared to their husbands? This is not known, but other studies in North Carolina farmwomen have suggested that in women who apply pesticides, risk was elevated in women who did not use protective clothing and gloves. Therefore, more information is needed on patterns of protection taken by both the wives and husbands to see if this influenced breast cancer risk associated with individual pesticides.

The study is based on 309 cases of breast cancer. While this is one of the largest studies of pesticide use and breast cancer risk, when looking at many individual pesticides, the number of breast cancer cases attributed to each of the 50 pesticides was often small. When the number of breast cancer cases associated with the use of a particular pesticide is small, the strength (the power) of the ability to detect an effect decreases. Therefore, these results should be considered preliminary. As more women develop breast cancer over the next five years, these added cases will give a better estimate of risk associated with specific pesticides.

Hoyer, A.P., Grandjean, P., Jorgensen, T., Brock, J.W., Hartvig, H.B. Organochlorine exposure and breast cancer risk, The Lancet, 352: 1816-1820, 1998.

Back to the top

Last Update 05.06.05