Pesticides [Updated April 2010]

Originally Posted April 2010 with contributions from ECWG members

INTRODUCTION

Pesticides comprise a large group of chemicals used as insecticides, herbicides and fungicides. Human exposure may occur: (a) in occupational settings such as during their industrial production and farming-associated use, (b) when used for cosmetic grass maintenance in homes, parks and golf courses, and (c) through ingestion of polluted water and food. Chemically, major classes of pesticides include organochlorines and organophosphates, important to distinguish as they may differ regarding their acute and chronic toxic properties in humans. Given the large diversity of pesticides, it is not possible or appropriate to generalize their health effects, including potential associations with cancer in humans. In this regard, research efforts have focused on specific chemicals that have showed carcinogenic properties in experimental models. There are many human studies addressing links between pesticide exposure and different cancers, the analysis of which is often limited by the large variety of human cancer types, the long period that may happen between exposure and cancer diagnosis, and the certainty that exposed humans are likely concurrently affected by many pesticides.

BEST EVIDENCE RESEARCH STUDIES

Using PubMed, we searched for all systematic reviews and meta-analytic studies that synthesized the many primary research studies addressing links between pesticide use and human cancer. The search strategy used included “pesticides” as a Medical Subject Heading (over 118,000 publications) and was limited only to “meta-analysis” resulting in only 147 publications. Screening of these allowed the identification of the studies listed below. The publications found were individually reviewed and jointly summarized by our Environment & Cancer Working Group. The following publications were appraised:

PESTICIDES AND LEUKEMIA

Turner et al., 2010. Residential pesticides and childhood leukemia: A systematic review and meta-analysis. Environmental Health Perspectives 118(1):33-41. This is the most updated study asking how pesticide exposure in children and mothers before and during pregnancy may affect the risk of childhood leukemia. It is based on 17 case-control studies that included leukemia in children and young adults up to 20 years of age and addressed outdoor and indoor exposure to pesticides. Exposure in all these studies was assessed only by surveying parents and not by measuring exposure. Childhood leukemia was significantly increased when pregnant women reported being exposed at home to pesticides (odds ratio 1.54, data from 11 studies), insecticides (odds ratio 2.05, data from 8 studies) and herbicides (odds ratio 1.61, data from 5 studies). These associations were stronger for children who developed acute lymphoblastic leukemia, when mothers reported to use pesticides themselves and for indoor use. Childhood exposure to pesticides, and specifically to insecticides, were also associated with increased risk of developing leukemia (odds ratios 1.38 and 1.61, respectively). [Summarized by Otto Sanchez].

Wigle et al., 2009. A systematic review and meta-analysis of childhood leukemia and parental occupational pesticide exposure. Environmental Health Perspectives 117(10):1505-1513. A study by the same Ottawa group as above (Turner et al., 2010) which complements it by focusing on parents working with pesticides and includes a total of 31 research studies. Overall, the authors did not find an association between occupational pesticide exposure to fathers and risk of leukemia in their children. However, when mothers were occupationally exposed to pesticides in farm-related activities before birth, there was an increase in risk of childhood leukemia (odds ratio 2.1-2.2). These associations were stronger for children who developed two common types of leukemia, namely, acute lymphoblastic leukemia (ALL; odds ratio 2.6) and acute myeloblastic leukemia (AML; odds ratio 2.6). [Reviewed by Otto Sanchez]

Van Maele-Fabry et al., 2008. Risk of leukaemia among pesticide manufacturing workers: A review and meta-analysis of cohort studies. Environmental Research 106:121-137. This meta-analysis examined the risk of leukemia associated with occupational exposure to pesticides, specifically focusing on workers involved in the manufacturing processes of these chemicals. It included 14 cohort studies from workers in various European countries and the United States of America. The results of the meta-analysis found that these pesticide manufacturing workers showed significantly increased risk of developing leukemia (1.43 ratio), especially those working with phenoxy herbicides (1.60 ratio). These associations were stronger for workers who developed leukemias of myeloid origin (6.99 ratio). [Reviewed by Jay Gill]

Van Maele-Fabry et al., 2007. A systematic review of myeloid leukemias and occupational pesticide exposure. Cancer Causes and Control 18:457-478. This article performed a systematic review and meta-analyses of 17 cohort and 16 case-control studies to determine if a potential association exists between myeloid leukemias (ML) and occupational pesticide exposure. In the end, for all cohort studies combined there was a slight (but not significant) overall increased risk for development of ML due to occupational pesticide exposure (Pooled rate ratio 1.21, 95% CI 0.99-1.48). The greatest risk for ML was found for individuals who worked as manufacturing workers (Rate ratio 6.32, 95% CI 1.90–21.01) and as pesticide applicators (Rate ratio 2.14, 95% CI 1.39-3.31). Associations were significantly stronger for workers who developed acute myeloid leukemia (Rate ratio 1.55, 95% CI 1.02-2.34). Inconsistencies in the overall results meant that the authors could not generate a solid conclusion as to whether or not a relationship existed between myeloid leukemia and overall occupational pesticide exposure. In the end, some evidence about the association between occupational pesticide and ML is brought forward in this article, but further study is needed to develop a more definitive answer and to generate greater evidence to support this association. [Reviewed by Jennifer Hepworth]

PESTICIDES AND LYMPHOMA

Merhi et al., 2007. Occupational exposure to pesticides and risk of hematopoietic cancers: Meta-analysis of case-control studies. Cancer Causes and Control 18:1209-1226. To our knowledge, this is the only study examining the risk of workers potentially exposed to pesticides to develop all types of hematopoietic cancers, including lymphomas, multiple myeloma, leukemias and myelodysplastic syndromes. The studies analyzed included workers diagnosed with any type of hematopoietic cancer between 1979 and 2003 in the United States of America, Australia, European countries and China. Synthesis of a total of 13 case-control studies revealed a positive and significant association between occupational exposure to pesticides and all hematopoietic cancer (1.33 odds ratio). Of all cancer types, the risk of developing Non-Hodgkin lymphoma was significantly increased (1.35 odds ratio). [Reviewed by Dana Thambiraj]

PESTICIDES AND BREAST CANCER

Khanjani et al., 2007. Systematic review and meta-analysis of cyclodiene insecticides and breast cancer. Journal of Environmental Science and Health Part C 25:23-52. This study focused on environmentally prevalent organochlorine insecticides known for their liposolubility, persistence in the human body particularly in fat-containing tissues, carcinogenicity by some in animal models and xenoestrogenic activity. It only included studies in which pesticides were measured in the blood or fat tissue of women with breast cancer. Meta-analysis of 21 case-control such studies, which analyzed ten pesticides and over 3,700 women with breast cancer, did not reveal significant associations between pesticide levels and breast cancer, except for heptachlor. For this particular insecticide, two studies from Italy and India including 305 women with breast cancer, together revealed significantly higher heptachlor levels in women with breast cancer than controls (geometric means ratio 5.32). [Reviewed by Otto Sanchez]

Lopez-Cervantes et al., 2004. Dichlorodiphenyldichloroethane burden and breast cancer risk: A meta-analysis of the epidemiologic evidence. Environmental Health Perspectives 112(2):207-214. The main metabolite of the insecticide dicholorodiphenyltricholoethane (DDT) is p, p’ –dichlorodiphenyldichloroethane (DDE). This meta-analysis included 22 case-control studies, published between 1993 and 2001, in which DDE levels in the serum and adipose tissue in over 4,500 women diagnosed with breast cancer were compared with women without breast cancer as controls. It included both retropective and prospective studies, as well as community- and hospital-based studies. Taken together, DDE levels did not correlate with development of breast cancer (summary odds ratio 0.97). This lack of correlation was not affected by the methodologies used to determine DDE levels, the specimen used or the type of case-control study. Despite the lack of positive association between p,p’ – DDE levels and breast cancer risk, the authors suggested that further research is necessary in areas such as the risk of developing adult breast cancer if exposed to DDT during the critical periods of development. [Reviewed by Sara Mannan]

PESTICIDES AND PROSTATE CANCER

Van Maele-Fabry et al., 2006. Review and meta-analysis of risk estimates for prostate cancer in pesticide manufacturing workers. Cancer Causes and Control 17:353-373. This meta-analysis synthesized studies addressing potential risks of developing prostate cancer for men who work in the pesticide manufacturing industry. It summarized original research studies published up to August 2004 and was based on a total of 18 cohort studies of workers in various European countries and the United States of America. Overall, it found a significantly increased risk of developing prostate cancer in pesticide manufacturing workers (odds ratio 1.28) and particularly in plants producing phenoxy herbicides (odds ratio 1.50). This important observation was limited by the lack of studies that accurately assess pesticide exposure in these workers. [Reviewed by Rima Karam]

Van Maele-Fabry and Willems, 2004. Prostate cancer among pesticide applicators: A meta-analysis. International Archives of Occupational and Environmental Health 77:559-570. This is a meta-analysis of seven case-control and fifteen cohort studies spanning a large time period, from 1966 to 2003, focusing on the risk of developing or dying from prostate cancer in occupational categories with a potentially large exposure to pesticides. Specifically, the authors identified nursery and greenhouse workers, pesticide applicators, farmers with a license to use pesticide, and employees of spraying companies as their occupational categories of interest. Relative risk (RR) estimates varied between 0.61 and 2.38, with nine estimates showing a negative association between the groups of interest and development of prostate cancer, and thirteen RR estimates reporting a positive association. Taken together, the studies revealed a meta-rate ratio of 1.24 which indicates an increase in risk of developing prostate cancer in the occupational categories of interest. Despite this observation, the authors advise against establishing causality between exposures to pesticides and prostate cancer in face of observed weak strength of associations, inconsistencies from study to study, a multi-factorial aetiology of prostate cancer, mixed results for dose-response relationships and a temporal relationship that cannot be fully assessed. [Reviewed by Irina Levit]

WHAT DO THESE STUDIES TELL US?

Exposure to pesticides is of concern to humans as these diverse chemicals can elicit acute and chronic toxic effects. Potential links between pesticide exposure and human cancer have been studied epidemiologically and experimentally. The research synthesis studies reviewed above focused on meta-analysis of case-control and cohort studies addressing correlations with childhood and adult leukemia, Non-Hodgkin lymphoma, breast and prostate cancer. Despite the fact that most of these studies do not specifically eliminate confounding factors and do not measure actual pesticide levels, residential exposure to pesticides appears to be linked with increased incidence of leukemia in children, especially when their pregnant mothers are exposed. As well, occupations in the pesticide manufacturing industry and involving pesticide application activities increased the risk of developing leukemia in their workers, particularly when working with phenoxy herbicides and showing a stronger association with myeloid leukemia types.

Occupational studies are important as they reflect effects in groups of workers with more homogeneous environments than the community at large, serve as sentinels for pesticides exposures in larger communities and show more obvious effects as workers are likely exposed to higher concentrations of undissolved chemicals than in the community. The studies reviewed above identified associations between occupational exposure to pesticides and development of Non-Hodgkin lymphoma and prostate cancer.

Lastly, breast cancer incidence was not epidemiologically associated with exposures to ten organochlorine pesticides or with measured levels of DDE, a primary metabolite of the insecticide DDT. The only exception was with heptachlor, in which meta-analysis of two studies including over 300 women with breast cancer revealed higher levels of this insecticide in women with breast cancer. We will continue updating this information as these results are challenged by new similar studies.

USEFUL ONLINE RESOURCES

Pesticides and Cancer – Canadian Cancer Society

Pesticides- California Department of Public Health

Pesticides- Environmental Protection Agency, USA

  1. No comments yet.

You must be logged in to post a comment.