Selection bias in case-control studies on household exposure to pesticides and childhood acute leukemia

Maternal and Paternal Household Pesticide Exposure During Pregnancy and Risk of Childhood Acute Lymphoblastic Leukemia

OBJECTIVE

The aim of this study was to investigate whether risk estimates for childhood acute lymphoblastic leukemia change when restricting model comparison groups to "nonpesticide exposure" (NPE10) households.

METHODS

Cases ( n = 1810) 15 years or younger were identified through Children's Cancer Group institutions between 1989 and 1993 and age-/sex-matched to controls ( n = 1951). Household pesticide use during pregnancy/month prior was collected via telephone. NPE10 comparison group reporting no parental exposure to 10 pesticide classes was identified.

RESULTS

Adjusted odds ratios increased from 15% to 49% when limiting the comparison to NPE10. Maternal termite insecticide exposure was associated with greatest risk (adjusted odds ratio, 4.21; 95% confidence interval, 2.00-8.88). There was minimal evidence of interaction by child sex or occupational pesticide exposure, and no monotonic dose-response pattern with frequency of use (times per year).

CONCLUSIONS

Elevated risks are consistent with published pooled-/meta-analyses and DNA damage. The consistency and magnitude of these associations warrant product labeling, exposure reduction interventions, or both.

Genome-wide association study reveals novel loci for adult type 1 diabetes in a 5-year nested case-control study

BACKGROUND

Type 1 diabetes (T1D) is a severe and prevalent metabolic disease. Due to its high heredity, an increasing number of genome-wide association studies have been performed, most of which were from hospital-based case-control studies with a relatively small sample size. The association of single nucleotide polymorphisms (SNPs) and T1D has been less studied and is less understood in natural cohorts.

AIM

To investigate the significant variants of T1D, which could be potential biomarkers for T1D prediction or even therapy.

METHODS

A genome-wide association study (GWAS) of adult T1D was performed in a nested case-control study (785 cases vs 804 controls) from a larger 5-year cohort study in Suzhou, China. Potential harmful or protective SNPs were evaluated for T1D. Subsequent expression and splicing quantitative trait loci (eQTL and sQTL) analyses were carried out to identify target genes modulated by these SNPs.

RESULTS

A harmful SNP for T1D, rs3117017 [odds ratio (OR) = 3.202, 95% confidence interval (CI): 2.296-4.466, P = 9.33 × 10^-4] and three protective SNPs rs55846421 (0.113, 0.081-0.156, 1.76 × 10^-9), rs75836320 (0.283, 0.205-0.392, 1.07 × 10^-4), rs362071 (0.568, 0.495-0.651, 1.66 × 10^-4) were identified. Twenty-two genes were further identified as potential candidates for T1D onset.

CONCLUSION

We identified a potential genetic basis of T1D, both protective and harmful, using a GWAS in a larger nested case-control study of a Chinese population.

Acute myeloid leukaemia at an early age: Reviewing the interaction between pesticide exposure and KMT2A-rearrangement

Acute myeloid leukaemia (AML) in early childhood is characterised by a high frequency of recurrent genomic aberrations associated with distinct myeloid subtypes, clinical outcomes and pathogenesis. Genomic instability is the first step of pathogenic mechanism in early childhood AML. A sum of adverse events is necessary to the development of infant AML (i-AML), which includes latency of biochemical-molecular and cellular effects. Inherited genetic susceptibility associated with exposures to biotransformation substances can modulate the risk of DNA damage and it is a very important piece in the pathogenic puzzle. In this review, we have aimed to explore the chain of events in the time-points of the natural history of i-AML, which includes maternal exposures during pregnancy, the speculations about the formation of somatic mutations during foetal life and the secondary genomic aberrations associated with i-AML. The modulation of risk conferred by xenobiotic metabolism´s genes variants is the bottom line of the pathogenic process. Since we have conducted observational and molecular investigations in early childhood leukaemia, the data focused here is based on Brazilian findings with summarised results of our experience with epidemiological and molecular studies in early-age leukaemia.

DNA methylation as a potential mediator of environmental risks in the development of childhood acute lymphoblastic leukemia

5-year survival rate for childhood acute lymphoblastic leukemia (ALL) has risen to approximately 90%, yet the causal disease pathway is still poorly understood. Evidence suggests multiple 'hits' are required for disease progression; an initial genetic abnormality followed by additional secondary 'hits'. It is plausible that environmental influences may trigger these secondary hits, and with the peak incidence of diagnosis between 2 and 5 years of age, early life exposures are likely to be key. DNA methylation can be modified by many environmental exposures and is dramatically altered in cancers, including childhood ALL. Here we explore the potential that DNA methylation may be involved in the causal pathway toward disease by acting as a mediator between established environmental factors and childhood ALL development.

Maternal Alcohol Consumption during Pregnancy and Early Age Leukemia Risk in Brazil

Objectives. To investigate the association between the maternal alcohol consumption during pregnancy and early age leukemia (EAL) in offspring. Methods. Datasets were analyzed from a case-control study carried out in Brazil during 1999-2007. Data were obtained by maternal interviews using a standardized questionnaire. The present study included 675 children (193 acute lymphoid leukemia (ALL), 59 acute myeloid leukemia (AML), and 423 controls). Unconditional logistic regression was performed, and adjusted odds ratios (adj. OR) on the association between alcohol consumption and EAL were ascertained. Results. Alcohol consumption was reported by 43% of ALL and 39% of AML case mothers and 35.5% of controls'. Beer consumption before and during pregnancy was associated with ALL in crude analysis (OR = 1.54, 95% CI, 1.08-2.19), although in adjusted analysis no statistical significance was found. For weekly intake of ≤1 glass (adj. OR = 1.30, 95% CI, 0.71-2.36) and ≥1 glass/week (adj. OR = 1.47, 95% CI, 0.88-2.46) a potential dose-response was observed (P trend < 0.03). Conclusion. This study failed to support the hypothesis of an increased risk of EAL associated with maternal alcohol intake during pregnancy, neither with the interaction with tobacco nor with alcohol consumption.

Bias--is it a problem, and what should we do?

Observational studies are prone to two types of errors: random and systematic. Random error arises as a result of variation between samples that might be drawn in a study and can be reduced by increasing the sample size. Systematic error arises from problems with the study design or the methods used to obtain the study data and is not influenced by sample size. Over the last 20 years, veterinary epidemiologists have made great progress in dealing more effectively with random error (particularly through the use of multilevel models) but paid relatively little attention to systematic error. Systematic errors can arise from unmeasured confounders, selection bias and information bias. Unmeasured confounders include both factors which are known to be confounders but which were not measured in a study and factors which are not known to be confounders. Confounders can bias results toward or away from the null. The impact of selection bias can also be difficult to predict and can be negligible or large. Although the direction of information bias is generally toward the null, this cannot be guaranteed and its impact might be very large. Methods of dealing with systematic errors include: qualitative assessment, quantitative bias analysis and incorporation of bias parameters into the statistical analyses.

Tobacco smoke exposure and the risk of childhood acute lymphoblastic and myeloid leukemias by cytogenetic subtype

BACKGROUND

Tobacco smoke contains carcinogens known to damage somatic and germ cells. We investigated the effect of tobacco smoke on the risk of childhood acute lymphoblastic leukemia (ALL) and myeloid leukemia (AML), especially subtypes of prenatal origin such as ALL with translocation t(12;21) or high-hyperdiploidy (51-67 chromosomes).

METHODS

We collected information on exposures to tobacco smoking before conception, during pregnancy, and after birth in 767 ALL cases, 135 AML cases, and 1,139 controls (1996-2008). Among cases, chromosome translocations, deletions, or aneuploidy were identified by conventional karyotype and fluorescence in situ hybridization.

RESULTS

Multivariable regression analyses for ALL and AML overall showed no definite evidence of associations with self-reported (yes/no) parental prenatal active smoking and child's passive smoking. However, children with history of paternal prenatal smoking combined with postnatal passive smoking had a 1.5-fold increased risk of ALL [95% confidence interval (CI), 1.01-2.23], compared to those without smoking history (ORs for pre- or postnatal smoking only were close to one). This joint effect was seen for B-cell precursor ALL with t(12;21) (OR = 2.08; 95% CI, 1.04-4.16), but not high hyperdiploid B-cell ALL. Similarly, child's passive smoking was associated with an elevated risk of AML with chromosome structural changes (OR = 2.76; 95% CI, 1.01-7.58), but not aneuploidy.

CONCLUSIONS

Our data suggest that exposure to tobacco smoking was associated with increased risks of childhood ALL and AML; and risks varied by timing of exposure (before and/or after birth) and cytogenetic subtype, based on imprecise estimates.

IMPACT

Parents should limit exposures to tobacco smoke before and after the child's birth.

Children's Oncology Group's 2013 blueprint for research: epidemiology

Investigators worldwide have for over 40 years conducted case-control studies aimed at determining the causes of childhood cancer. The central challenge to conducting such research is the rarity of childhood cancer, thus many studies aggregate cases through clinical trials organizations such as COG. Rarity also precludes the use of prospective study designs, which are less prone to recall and selection biases. Despite these challenges a substantial literature on childhood cancer etiology has emerged but few strong environmental risk factors have been identified. Genetic studies are thus now coming to the fore with some success. The ultimate aim of epidemiologic studies is to reduce the population burden of childhood cancer by suggesting preventive measures or possibly by enabling early detection. Pediatr Blood Cancer 2013; 60: 1059-1062. © 2012 Wiley Periodicals, Inc.

In utero pesticide exposure and leukemia in Brazilian children < 2 years of age

BACKGROUND

An association between pesticide exposure and cancer has been suggested. Infant leukemia is a rare neoplasm and its association with maternal pesticide exposure has been poorly explored.

OBJECTIVES

We investigated the association between pesticide exposure during pregnancy and leukemia in children < 2 years of age.

METHODS

A hospital-based case-control study was carried out in 13 Brazilian states during 1999-2007. Mothers of 252 cases and those of 423 controls were interviewed. Information on pesticide exposures 3 months before pregnancy, throughout pregnancy, and during breastfeeding was obtained. Unconditional logistic regression was used to estimate adjusted odds ratios (aORs) for associations between pesticide exposures and leukemia.

RESULTS

Associations with ever use of pesticides during pregnancy were observed for acute lymphoid leukemia (ALL) (aOR = 2.10; 95% CI: 1.14, 3.86) and acute myeloid leukemia (AML) (aOR = 5.01; 95% CI: 1.97, 12.7) in children 0-11 months of age, and with ALL (aOR = 1.88; 95% CI: 1.05, 5.23) at 12-23 months of age. According to reported maternal exposure to permethrin, higher risk estimates were verified for children 0-11 months of age (aOR = 2.47; 95% CI: 1.17, 5.25 for ALL; and aOR = 7.28; 95% CI: 2.60, 20.38 for AML). Maternal pesticide exposure related to agricultural activities showed an aOR of 5.25 (95% CI: 1.83, 15.08) for ALL, and an aOR of 7.56 (95% CI: 1.83, 31.23) for AML.

CONCLUSIONS

These results support the hypothesis that pesticide exposure during pregnancy may be involved in the etiology of acute leukemia in children < 2 years of age.

Residential exposure to pesticides and childhood leukaemia: a systematic review and meta-analysis

OBJECTIVE

To conduct a systematic review of published studies on the association between residential/household/domestic exposure to pesticides and childhood leukaemia, and to provide a quantitative estimate of the risk.

METHODS

Publications in English were searched in MEDLINE (1966-31 December 2009) and from the reference list of identified publications. Extraction of relative risk (RR) estimates was performed independently by 2 authors using predefined inclusion criteria. Meta-rate ratio estimates (mRR) were calculated according to fixed and random-effect models. Separate analyses were conducted after stratification for exposure time windows, residential exposure location, biocide category and type of leukaemia.

RESULTS

RR estimates were extracted from 13 case-control studies published between 1987 and 2009. Statistically significant associations with childhood leukaemia were observed when combining all studies (mRR: 1.74, 95% CI: 1.37-2.21). Exposure during and after pregnancy was positively associated with childhood leukaemia, with the strongest risk for exposure during pregnancy (mRR: 2.19, 95% CI: 1.92-2.50). Other stratifications showed the greatest risk estimates for indoor exposure (mRR: 1.74, 95% CI: 1.45-2.09), for exposure to insecticides (mRR: 1.73, 95% CI: 1.33-2.26) as well as for acute non-lymphocytic leukaemia (ANLL) (mRR: 2.30, 95% CI: 1.53-3.45). Outdoor exposure and exposure of children to herbicides (after pregnancy) were not significantly associated with childhood leukaemia (mRR: 1.21, 95% CI: 0.97-1.52; mRR: 1.16, 95% CI: 0.76-1.76, respectively).

CONCLUSIONS

Our findings support the assumption that residential pesticide exposure may be a contributing risk factor for childhood leukaemia but available data were too scarce for causality ascertainment. It may be opportune to consider preventive actions, including educational measures, to decrease the use of pesticides for residential purposes and particularly the use of indoor insecticides during pregnancy.