Pesticides, other chemical exposures and risk of childhood leukemia

ehasl research project Childhood leukemia is the most common type of childhood cancer and its etiology is largely unknown. Established risk factors account for only 10% of childhood leukemia incidence. Investigators at the University of California at Berkeley (UCB) conducted a case-control study of childhood leukemia from mid-1995 through 1999 in 17 counties in the San Francisco Bay area. They received additional funding to extend the study four years and to expand the study area to 18 additional counties in the Central Valley of California. Advantages of this extension will be to increase the sample size and to increase the proportion of Hispanics, who experience higher rates of childhood leukemia compared with non-Hispanic whites, and among whom the incidence of childhood leukemia is increasing. Another advantage of the extension is that a significant percent of the population in the Central Valley is employed in agriculture or lives in close proximity to agricultural fields, making this an ideal study population to evaluate agricultural pesticide exposures and risk of childhood leukemia. No previous large-scale childhood leukemia studies have targeted Hispanics or rural and agricultural populations.

Current evidence suggests that leukemia results from chromosome alterations and mutations that disrupt the normal process by which lymphoid or myeloid progenitor cells differentiate. Important molecular changes may be inherited at conception, or may occur during fetal development, during infancy, or early childhood. The UCB study will collect pretreatment blood and bone marrow in order to characterize specific translocations and mutations.

A major focus of the UCB study is to evaluate whether exposure to pesticides increases the risk of childhood leukemia. Home and garden pesticide use and parental occupational exposure have been associated with an increased risk of childhood leukemia in many epidemiologic studies, although most previous studies have not obtained information about specific active ingredients or about the timing of exposure. In the current study design, pesticide use information is obtained by in-person interviews and an inventory of pesticides in the home. The home inventory of pesticides will allow for the ascertainment of specific pesticide active ingredients for home and garden pesticides and will allow for the evaluation of the reliability of self-reported pesticide use. Occupational exposures for both the mother and the father in the year prior to conception, during pregnancy, and in the first years of the child's life will be obtained through a complete work history. Detailed information on potential pesticide exposures will be obtained through the use of specific job modules for farmworkers, farmers, pesticide applicators, gardeners, greenhouse workers, and fruit and vegetable packers.

We are currently conducting a study to further improve upon the pesticide exposure assessment of the University of California Berkeley study by providing additional funding for the measurement of pesticide levels in carpet dust in the homes, and for the mapping of residences to determine their proximity to agricultural fields treated with pesticides.

Carpet dust can be a reservoir for pesticides and other chemicals because they are protected from degradation by microbial action, moisture and sunlight. Pesticides found in carpet dust include those recently used around the home and on nearby crop fields. Pesticides used many decades ago are also frequently detected. Other commonly detected chemicals in carpet dust include polycyclic aromatic hydrocarbons (PAHs) and polychlorinated biphenyls (PCBs).

The pesticides detected in carpet dust depend on the persistence and volatility of the chemicals. The detection of persistent organochlorine insecticides (e.g., DDT) that were used years ago indicate that residential exposure was likely to have occurred over many years, including the time period relevant for childhood leukemia. However, many pesticides measured in carpet dust at the time of diagnosis may only reflect current use in or near the home. To provide historical exposure information for less persistent pesticides, the UCB childhood leukemia study, like previous epidemiologic studies, includes detailed questions about home, garden, and occupational use of pesticides. However, questionnaires are not useful for determining indirect exposure to agricultural pesticides used near the home. This is an important exposure assessment issue in our study, because the Central Valley of California has one of the highest reported rates of use of agricultural pesticides in the USA. Due to the limitations of current approaches for estimating indirect exposures to agricultural pesticides, it is important to determine if other information can be used to estimate past agricultural pesticide exposures during time periods relevant for cancer risk. A method has been developed for identifying residences with probable exposure from agricultural pesticide drift. This method makes use of remotely sensed data and a geographic information system (GIS) to identify specific crops grown in the vicinity of residences. A modification of this method has been developed for California agricultural. The methods uses land use data to map the location of crop fields near residences. Linkage of these maps with crop-specific pesticide use data results in an exposure metric for estimating indirect exposures to agricultural pesticides for each residence. In collaboration with California Department of Health Services (CDOH) researchers, these exposure metrics were determined for 852 residences from a case-control study in the Central Valley of California.