The Pediatric Page
Childhood Cancer and Environmental Exposures (May 2005)
Environmental exposures at home, work and in our communities are front-page news. In an era of allergen-free spaces, organic food and non-synthetic materials, parents constantly ask about their significance in the etiology of pediatric cancer and other chronic illness.
Today, 78% of all children with cancer are expected to be cured. However, cancer is still the 2nd to 4th leading cause of death (depending on age group) among children older than one(1). That is why understanding cancer’s etiology occupies an ever growing segment of pediatric oncology research.
Epidemiologic studies have led to major advances as in smoking and lung cancer, fiber and colon cancer. However, the design and interpretation of a study can create misconceptions (such as the relationship of cancer and electromagnetic fields – initially thought to induce cancer). Two different types of studies can be performed to identify and quantify the power of an exposure: cohort studies (comparing an exposed and unexposed population) and case-control studies (population with the disease is compared to a random group of people without the disease, looking for possible exposures). The association of the exposure to the occurrence of cancer is expressed in odds ratio (OR = observed over expected cases). Both types of studies are difficult and mostly retrospective. Moreover, since childhood cancer is not common, very large cohort studies must be performed to prove significance. Population studies in the U.S. are not feasible because of the lack of centralized data banking. As a result, studies must be case-controlled and are subject to bias such as: non-balance by income, ethnicity or even age. Although the power of these studies is limited, they can lead to breakthroughs in public health and biological advances.
Below we list specific environmental factors and their known, limited or suggested association with cancer. Some of these studies are ongoing and will become clear within the next five years. Some deserve further discussion.
Childhood Cancer and Environmental Exposure (summary)
Cancer Subtype: Leukemias (ALL)
Known Association: Ionizing radiation - in utero diagnostic, therapeutic postnatal
Suggested Association: Household pesticide use
Limited Association: Parental exposure - hydrocarbons, paints, vehicle exhaust, electromagnetic fields (?)
Cancer Subtype: Leukemias (AML)
Known Association: Ionizing radiation - therapeutic postnatal
Suggested Association: Maternal alcohol use, parental exposure - benzene, pesticides
Limited Association: Maternal use of marijuana use during pregnancy; parental occupational exposure: pesticides
Cancer Subtype: Brain Tumors
Known Association: Ionizing radiation
Suggested Association: Maternal diet - cured meats
Limited Association: Parental occupation - electronics, petrol, pulp, metal, solvents, residential pesticides
Cancer Subtype: Hodgkin's Lymphoma
Known Association: Infection - EBV
Suggested Association: Residential exposure - pesticides
Limited Association: -
Cancer Subtype: Non-Hodgkin's Lymphoma
Known Association: -
Suggested Association: -
Limited Association: -
Cancer Subtype: Bone Tumors
Known Association: Radiation therapy for childhood cancer, alkylating agents
Suggested Association: -
Limited Association: Parental occupation: chicken farming, pesticides
Cancer Subtype: Soft Tissue Sarcomas
Known Association: -
Suggested Association: -
Limited Association: Diagnostic radiographs during pregnancy; parents' use of recreational drugs
Cancer Subtype: Renal Tumors
Known Association: -
Suggested Association: Parental occupation - welding, mechanic
Limited Association: Parental exposure - pesticides; maternal use - coffee or tea during pregnancy, hair dye
In utero diagnostic X-ray (pelvimetry) is mildly associated with childhood cancer, mostly in children with acute lymphoblastic leukemia (ALL). However, the use of pelvimetry during the third trimester has diminished significantly during the last 15 years. This is reflected in the higher prevalence of ALL in exposed older children, ages 11-15, but not younger children. Also, the better prognostic subtype of ALL, early pre-B, is less prevalent in patients exposed to x-rays. Diagnostic ultrasound is not associated with pediatric cancer(2).
There is a suggestive association of hydrocarbon exposure (automobile traffic), measured by the distance of the home from the highway with an odd ratio of 8.28 for leukemia(3).
The population-based studies in Norway and Great Britain have found no association of childhood cancer and specifically CNS tumors to electromagnetic fields. This included: electricity supply equipment, distances to high-voltage lines, underground cables(4).
Pesticide exposure at home and through parents who handle pesticides have been studied(5,6). Although the results are early and studies must be continued, there is an increased risk of cancer (highest in lymphomas). OR was 1.98. The use of gloves decreased the risk. The Northern California Leukemia Study (NCLS) has released their preliminary results of a case-control study of 847 children with leukemia. They found that the use of household pesticides (especially during pregnancy) correlate with an increased risk of acute lymphoblastic leukemia(7). Similar findings have been reported in Germany(8).
Exploring the role of infectious disease in Hodgkin’s, researchers found that attendance at nursery school or daycare for at least one year protected against the lymphoma(9). Similar findings are reported by the NCLS.
Infant ALL and AML are associated with chromosomal rearrangement involving the MLL gene at chromosome 11q23. MLL gene rearrangements are found in people with secondary AML after exposure to DNA topoisomerase II inhibitors (etoposide and anthracyclines). There are natural DNA topoisomerase II inhibitors in foods: the flavinoids in vegetables and some fruits. Studies of maternal diet found that a high consumption of flavinoids (highest in onions, apples and canned beans) had an OR of 1.98 for MLL positive myeloid leukemias (not lymphoid leukemia). However, a maternal diet high in vegetables protected against all childhood leukemias(10). These findings suggest that specific genotypes may result in delayed detoxification of specific noxa and therefore a higher risk of cancer. For example: NAD(P)H detoxifies the flavinoid quinones. Specific genotypes of NAD(P)H:quinone oxidoreductase are associated with higher incidence of MLL positive infant leukemias(11). This type of biological finding can help epidemiologists find associations between cancer and exposures. At the same time epidemiology helps to reveal novel mechanistic etiologies that may prove important in the near future.
References
- Brent RL, Weitzman M. The pediatrician’s role and responsibility in educating parents about environmental risks. Pediatrics, 113:1167-72, 2004
- Linet MS, et al. Interpreting epidemiologic research: lessons from studies of childhood cancer. Pediatrics, 112: 218-232, 2003.
- Shu XO, et al. Diagnostic S ray and ultrasound exposure and risk of childhood acute Lymphoblastic leukemia by immunophenotype. Ca, Epidem, Biomar Preven, 11, 177-185, 2002
- Pearson RL. Distance-weighted traffic density in proximity to a home is a risk factor for leukemia and other childhood cancers. J Air Waste Manag Ass, 50(2):175-180, 2000.
- Anonymous. Childhood cancer and residential proximity to power lines. UK Childhood Study Investigators. British J of Ca, 83(11):1573-80, 2000
- Flower KB et al. Cancer risk and parental pesticide application in children of agricultural health study participants. J of the National Institute of Environmental Health Sciences, online, Dec 2003.
- Metayer, C. California Childhood Leukemia Study: 10 years of experience in environmental and genetic epidemiology. Childrens Cancer Group Epidemiology Symposium, April 2005.
- Meinert R et al. Leukemia and non-Hodgkin’s lymphoma in childhood and exposure to pesticides: results of a register-based case-control study in Germany. Ameri J of Epidemiol, 151(7): 637-50, 2000.
- Chang ET et al. Childhood social environment and Hodgkin’s lymphoma: new findings from a population-based case control study. Ca Epidemol Bioma Prev 13(8): 1361-70, 2004.
- Spector LG et al. Maternal diet and infant leukemia: the DNA topoisomerase II inhibitor hypothesis: a report from the Children’s Oncology Group. Ca Epidemiol Biomark Prev, 14(3): 651-5, 2005.
- Wiemels JL et al. A lack of a functional NAD(P)H: quinone oxidoreductase allele is selectively associated with pediatric leukemias that have MLL fusions. Ca Res, 59, 4095-99, 1999.
This information provided by Sandra Luna-Fineman, M.D., Pediatric Hematology and Oncology Program
Tel: (415) 600-3477