2019 Leon

Original Article Overview

  • Title: Pesticide use and risk of non-Hodgkin lymphoid malignancies in agricultural cohorts from France, Norway and the USA: a pooled analysis from the AGRICOH consortium

  • Authors:

    • Maria E Leon

    • Leah H Schinasi

    • Pierre Lebailly

    • Laura E Beane Freeman

    • Karl-Christian Nordby

    • Gilles Ferro

    • Alain Monnereau

    • Maartje Brouwer

    • Séverine Tual

    • Isabelle Baldi

    • Kristina Kjaerheim

    • Jonathan N Hofmann

    • Petter Kristensen

    • Stella Koutros

    • Kurt Straif

    • Hans Kromhout

    • Joachim Schüz

  • Correspondence: Maria E Leon, International Agency for Research on Cancer (IARC), Lyon, France.

  • Editorial Decision: 22 January 2019; Accepted: 4 February 2019

Abstract

Background

  • Pesticide utilization in agriculture is common. Previous studies have indicated a need for detailed investigations into potential associations between pesticide exposure and lymphoid malignancies among agricultural workers.

Methods

  • Evaluation focused on 14 pesticide chemical groups and 33 individual active ingredients related to non-Hodgkin lymphoid malignancies (NHL).

  • Data Sources:

    • Self-reported pesticide use history combined with crop-exposure matrices (France and Norway).

    • Self-reported lifetime pesticide use (USA).

  • Analytical Method: Cox regression models used for estimating cohort-specific hazard ratios (HRs) and 95% confidence intervals (CIs), combined via random effects meta-analysis.

Results

  • 2430 NHL cases diagnosed among 316,270 farmers over 3,574,815 person-years.

  • **Initial Findings: ** Most meta-HRs suggested no association between pesticide use and NHL.

  • Moderately elevated meta-HRs:

    • NHL with terbufos: meta-HR = 1.18 (95% CI: 1.00–1.39)

    • Chronic lymphocytic leukaemia/small lymphocytic lymphoma (CLL/SLL) with deltamethrin: meta-HR = 1.48 (95% CI: 1.06–2.07)

    • Diffuse large B-cell lymphoma (DLBCL) with glyphosate: meta-HR = 1.36 (95% CI: 1.00–1.85)

  • Inverse associations for NHL:

    • Organochlorine insecticides: meta-HR = 0.86 (95% CI: 0.74–0.99)

    • Phenoxy herbicides: meta-HR = 0.81 (95% CI: 0.67–0.98)

  • Important Limitations: Non-differential exposure misclassification necessitating improved exposure estimates.

Key Terms

  • Pesticides: Chemicals used in agriculture to manage pests.

  • Non-Hodgkin lymphoid malignancies (NHL): A diverse group of blood cancers originating in the lymphatic system.

Introduction

  • Pesticide exposure is prevalent in agriculture, especially among farmers engaged in various tasks contributing to exposure.

  • Tasks include:

    • Mixing/loading chemicals

    • Applying pesticides in fields

    • Treating seeds

    • Handling and cleaning equipment

  • Biological effects potentially caused by pesticide exposure include genotoxicity, immunosuppression, oxidative stress, and hormonal alterations, which can be indicative of carcinogenic properties.

  • Chronic exposure could partly account for higher cancer incidence rates in farmers.

AGRICOH Consortium

  • The AGRICOH consortium includes multiple agricultural cohorts investigating the link between pesticide exposure and NHL.

  • The current study synthesizes data from three substantial cohorts to analyze the connection between selected pesticide groups and NHL.

Methods

Study Population

  • Description of participating cohorts:

1. Agriculture and Cancer (AGRICAN)

  • Enrolled 181,747 individuals aged ≥18 within the French agricultural health insurance system from 2005 to 2007.

  • Participants completed questionnaires about crops cultivated and use of pesticides.

  • Linkage to cancer registries was performed until December 2009.

2. Cancer in Norwegian Agricultural Population (CNAP)

  • Comprises farm holders and their families included in various national census data collected between 1969 and 1989.

  • Included 147,134 individuals assessed for crop and pesticide usage linked to the Cancer Registry until December 2011.

3. Agricultural Health Study (AHS)

  • Enrolled 52,394 private pesticide applicators and included follow-up questionnaires about pesticide use and practices from 1993 to 1997.

  • Cohorts linked to cancer registries until December 2011.

Exposure Assessment

  • Cross-referencing cohorts with country-specific crop-exposure matrices (CEMs) to ascertain potential exposure:

    • Usage pattern: derived from crop cultivation reports and treatment tasks.

  • Participants in the AHS reported their pesticide applied, creating a self-reported exposure evaluation.

Results

Population Characteristics

  • Total participants: 316,270

    • Gender Distribution: 237,317 males (75%), 78,953 females (25%).

  • The follow-up period ranged from 1993 to 2011, with analyses capturing 2430 NHL cases identified.

  • Median age at diagnosis was 69 years, with cohorts varying in age and exposure prevalence:

- AGRICAN Characteristics

  • Males: 56% (71,358)

  • Females: 44% (55,924)

  • Median age: 67

  • Exposure to any pesticides: 68%

- CNAP Characteristics

  • Males: 84% (116,128)

  • Females: 16% (21,693)

  • Median age: 51

  • Exposure to any pesticides: 63%

- AHS Characteristics

  • Males: 97% (49,831)

  • Females: 3% (1,336)

  • Median age: 46

  • Exposure to any pesticides: 99%

Statistical Analysis

  • Cox proportional hazard models were utilized to calculate hazard ratios with 95% confidence intervals for incidences of NHL relating to pesticide use.

Discussion

Summary of Key Findings

  • It is crucial to emphasize the differing associations of NHL risk with specific chemical groups and assessment of association strength through meta-analysis.

  • Associations observed for:

  • Terbufos (HR = 1.18)

  • Deltamethrin (CLL/SLL, HR = 1.48)

  • Glyphosate (DLBCL, HR = 1.36)

  • Inverse associations for organochlorine insecticides and phenoxy herbicides.

Implications of Findings

  • The study suggests a need for further investigations into subtype-specific associations and enhancing methodology for exposure assessment to facilitate clearer conclusions. Future research should refine methodology with considerations of duration, frequency, and intensity of exposure to pesticides, subsequently contributing to our understanding of occupational hazard impacts in agricultural settings.

Study Limitations

  • Non-differential exposure misclassification potentially attenuates observed relationships, potentially underestimating the risk of NHL.

  • Cross-cohort heterogeneity may distort risk association evaluations.

Conclusion

  • The analysis brings forward insights into specific pesticide associations with NHL, advocating for broader studies to enhance the understanding of the link between pesticide exposure, resulting in recommendations for refined methodologies in future investigations.

  • Continued international collaborations on research surrounding the carcinogenic effects of pesticide exposure remain essential to uncover further nuanced findings from agricultural health studies.

Supplementary Data

  • Available via the International Journal of Epidemiology online portal. Numerous references document varying phenomena related to pesticide exposure and health risk analysis, reflecting on studies core to understanding the broader impacts: studies referenced include papers analyzing non-Hodgkin lymphoma risks associated with specific pesticides and investigations into environmental health perspectives.

Funding

  • Supported by ONEMA, IARC, and additional contributions affirmed by various research grants.