Michael Alavanja, Dr.P.H. Captain, USPHS Senior Investigator, Division of Cancer Epidemiology and Ge

1
Michael Alavanja, Dr.P.H.Captain, USPHSSenior
Investigator, Division of Cancer Epidemiology
and Genetics, NCI2007 North American Pesticide
ApplicatorCertification Safety Education
WorkshopAugust 20-23, 2007Portland, Maine
2
Agricultural Health Study on Cancer Findings.
Session I General StrategyTuesday, August
21Breakout Session 1

3
Do pesticides cause cancer ?

• Few strong and consistent associations linking a
  single chemical to a single cancer.
• Animal/laboratory studies show most pesticides in
  current use to be non-genotoxic.
• Exposure assessment in previous epidemiologic
  studies was general weak, they were based on
  interviews and could suffer from case recall
  bias.
• Studies of pesticide manufactures are generally
  too small to give meaningful results for cancer
• Exposures among the general population in
  developed countries are relative low and effect
  hard to measure.
• In summary Neither animal studies nor human
  studies give a compelling case for an
  association.

4
Background

• World-wide occupational exposures to pesticides
  exceed 1.8 billion people (World Bank estimate).
• Everyone in the USA has some indirect exposure to
  pesticides (NHANES).
• Agricultural Insecticides as a group labeled as
  probable (group 2A) human carcinogens by IARC.
• Only arsenic and dioxin are listed as human
  carcinogens by IARC.
• Vital public health need to identify human
  carcinogens on the market!

5
Background

• The Occupational Environmental Epidemiology
  Branch, NCI has a history of ecological and
  case-control of farmers starting in the 1970s.
• A common critique-? exposure assessment was weak.
• I proposed the idea for a prospective cohort
  study of pesticide applicators in 1989- 1990.
• In 1991 an extramural advisory group
  recommended the OES conduct the AHS.
• The Agricultural Heath Study entered the field in
  December 12, 1993.
• Other federal partners joined the team in 1994
  (EPA), 1995 (NIEHS) and NIOSH (1997).

6
Design AHS(www.aghealth.org)

• Prospective cohort study of 89,658 pesticide
  applicators spouses (IA and NC).
• 82 of target population enrolled 1993-1997.
• Little loss to follow-up (lt2).
• Cancer incidence and mortality updated annually.
• Comprehensive exposure assessment information on
  82 pesticides collected at three points in time.
• Questionnaire exposure assessment evaluated with
  field measurements of pesticides.
• Buccal cells collected on gt35,000 study subjects.

7
Disease Etiology In the AHS

• Central Research Objectives
• 1. Characterize exposures to the highest degree
  ever achieved in large cohort study.
• 2. Identify pesticides and other agricultural
  exposures that increase the risk of cancer .
• 3. Identify the mode of action of agents
  causing disease.

8
Types of Pesticide Exposure

• Acute exposure events. High exposure dose, short
  time period (minutes or hours).
• Chronic exposure. Low exposure dose, long time
  period (hundreds or thousands of days in a
  lifetime).

9
Agricultural Health StudyPesticide Exposure
Estimates
Calculating Cumulative Exposure Index
Cumulative Exposure Intensity Duration
Where Intensity Exposure scores obtained
from algorithms Duration Days/years
Years/life-time days/life-time From
Dosemeci et al. Ann Occup Hyg 46245-260, 2002.
10
(No Transcript)
11
Questionnaire Evaluation Monitoring Visits
12
Questionnaire Evaluated with Field Measurements
of 2,4-D and Other Pesticides
Technician observations MLA
Questionnaire

Day 1 Day 2 Day
34

• Mix Load Apply (MLA).
• Hand wipes after MLA
• Dermal patches
• Air measurements

Collect full first morning void
3. Collect each void from MLA through next
morning void.
Collect full first morning void
13
Comparison of Questionnaire Based Intensity
Scores and Field Measurements 2,4-D(Thomas et
al., in review)
14
Conclusions From Exposure Algorithm Assessment

• For 2,4-D applicators we observed a significant
  correlation between the questionnaire-based
  algorithm (intensity-factor) and post-application
  urine concentrations.
• Important additional determinants of exposure
  have been identified to refine the exposure
  algorithm.

15

• Evaluating the association between estimated
  exposures with health effects.
• (Cancer Etiology Studies in the AHS)

16
Effect

• End point of a causal mechanism.
• Amount of change in a populations disease
  frequency caused by a specific factor.
• Incident rate Number of new cases of disease in
  a specified period of time.
• Absolute effect I1 I0
• Relative Effect I1 / I0

17
Confounding factors

• A confounding factor must be a risk factor for
  the disease.
• A confounding factor must be associated with the
  exposure under study in the source population
  (the population from which the cases are
  derived).
• A confounding factor must not be effected by the
  exposure or the disease. In particular, it cannot
  be an intermediate step in the causal path
  between the exposure and the disease.
• How do we control confounding? Collect
  quantitative information on the exposure to the
  confounder and add the term to the multivariate
  model yb0 b1x1 b2 x2

18
Statistical Interaction-Effect Modification

• An effect-modifier is an exposure or host factor
  that modulates the extent of the effect of the
  study variable on the disease under
  investigation.
• If a cohort is divided into two or more distinct
  categories defined by the level of an effect
  modifier the stratum-specific effect measures may
  or may not be equal. If they are equal there is
  no effect modification. If they are significantly
  different there is effect modification.
• How do measure effect modification? Collect
  quantitative information on the exposure thought
  to be an effect modifier and add the product term
  to the multivariate model yb0 b1x1 b2 x2
  b 3 x1 x2

19
Typical Sequence of Cancer Etiology Studies in
AHS 2003-2007

• SIR analysis (generates general hypothesis) n1
• Nested case-control study of specific cancers
  (generate specific hypotheses n6)
• 1 ST COHORT ANALYSIS of specific pesticide
  (generates or refines specific hypotheses n21)
• 2ND COHORT ANALYSIS
• (Test Specific Hypotheses n1 in progress)
• Molecular epidemiology studies of cancer
  (Evaluates biological plausibility and mode of
  action n3 in progress)

20
AHS Research StrategyMitigate False Positive
Results
Biological Initial Replication Evidence
in Findings later in time Humans
Iowa North Carolina License Type
Exposure- Response Exposure-
Response Exposure- Response
Exposure- Response Exposure-
Response Exposure- Response
YES YES YES
21
Mitigate False-Positive Associations and Study
Rare Diseases

• Agricultural Health Cohort Consortium.
• (NCI organized)

22
Regulatory Implications of AHS Findings

• International Agency for Research on Cancer.
• International recommendations
• United States Environmental Protection Agency
• Educating pesticide applicators
• Label instructions
• Limitations of use
• Banning use

23
Thank you for listening AHS Research Team