Is Chemical Pollution Responsible for Childhood Tumors? : Epidemiology

Secondary Logo

Journal Logo


Is Chemical Pollution Responsible for Childhood Tumors?

Petridou, Eleni

Author Information
  • Free

In this issue of Epidemiology, two papers provide evidence that chemical pollutants could be responsible for some fraction of the occurrence of childhood leukemia 1 and neuroblastoma. 2 Childhood tumors are different from adult onset cancers in several ways, one of which is particularly challenging to epidemiologists: whereas for several adult-onset tumors, epidemiology and prevention have made substantially more contribution than treatment, the opposite is true for childhood tumors. For these neoplasms epidemiology and prevention have contributed little, in striking contrast to the remarkable therapeutic improvements during the last few decades for many forms of childhood malignancies, notably for childhood leukemia. 3

Childhood Leukemia

The vast majority of childhood leukemia is acute lymphoblastic leukemia, for which there is some evidence that the incidence may be increasing slightly. 4 Little is known about its etiology except that genetic factors play a role 5 and that ionizing radiation weights more heavily in the etiology of acute lymphoblastic leukemia than in that of most other malignancies. 6,7 These causes, however, account for only a very small fraction of cases of childhood leukemia. The hypothesis that exposure to extremely low frequency electric and/or magnetic fields is responsible for a large fraction of acute lymphoblastic leukemia has only weak empirical support. 8–11 Thus, the etiology of the large majority of childhood leukemia remains unexplained. Two general hypotheses have competed for the vacuum: one focuses on viral infections, the other on chemical environmental exposures.

The central role that viruses play in leukemogenesis in several animal species and the documentation of the human T-cell leukemia/lymphoma virus in a very rare form of leukemia have lead plausibility to the hypothesis of viral leukemogenesis in children. On the other hand, there has been no laboratory support for the hypothesis, and several investigators have postulated that childhood leukemia may be only a rare outcome of a common infection in a background of low herd immunity. 12–15 In contrast, there is no theoretical undermining for a role of chemical environmental pollution in the causation of acute lymphoblastic leukemia, and this relation has not been investigated through analytical, as opposed to ecological designs. 16 The paper by Infante-Rivard et al. in this issue 1 presents results for what may well be the most sophisticated epidemiologic investigation to date of acute lymphoblastic leukemia in relation to drinking water contamination.

The study by Infante-Rivard et al.1 is a large population-based case-control investigation that was undertaken with an elaborate protocol to evaluate the relation between childhood leukemia and drinking water contaminants, specifically total and selected trihalomethanes, certain metals, and nitrates. The authors developed an exposure matrix on the basis of municipal and provincial historical data and a tap water survey. None of these sources was complete with respect to any of the studied exposures, and imputations and occasionally arbitrary choices were necessary. Nevertheless, it is hard to think of a better design or a more satisfactory context for such an investigation. The authors evaluated average exposure level as well as cumulative exposure and they have focused on both the pregnancy and the postnatal period. They concluded that “the indications for an association between childhood leukemia and disinfection by-products as well as some metals are not strong, nor are they absent, in particular for postnatal exposure.”

Despite the expertise and the amount of work the authors invested in this study, I do not agree completely with their conclusion. In my opinion, the study provides very little evidence for any association between the studied exposures and childhood leukemia. The authors report that there are no important differences with respect to average values for any of the studied exposures in either the prenatal or in the postnatal period. They consider notable, however, the apparent excess risk for acute lymphoblastic leukemia among children postnatally exposed to cumulative levels of total trihalomethanes, in particular, chloroform above the 95th percentile, even though the increases are trivial; they relate to a small proportion of children and they could even be due to differences in duration of exposure, generated by unavoidably suboptimal age-matching of cases and controls. The authors also consider as noteworthy the excess risk for acute lymphoblastic leukemia in relation to cumulative zinc levels above the 95th percentile, even though the evidence for carcinogenicity of zinc is generally minimal. Regardless of the interpretation, it is clear that the evaluated contaminants of drinking water can explain no more than a trivial fraction of the total cases of acute lymphoblastic leukemia, and possibly none at all.


Neuroblastoma is a rare childhood tumor and yet it is the most common tumor in the 1st year of life. It derives from embryonal cells in the neural crest, and it arises in the adrenal medulla or anywhere else in the sympathetic chain. The disease appears to be less common in developing countries and among preterm babies. Neuroblastoma is accompanied by fever and weight loss; the physical examination reveals an abdominal mass. Diagnosis relies on ultrasound, computerized tomography, excretion of catecholamines in the urine and, eventually, biopsy. The prognosis of the disease is good when the tumor is detected before the first year but it is poor when the tumor is detected later in life. The aberrant expression of the MYCN oncogene is considered a marker of poor prognosis.

Few studies have evaluated the etiology of neuroblastoma. Exposure to pesticides was considered in some of them 17–21 and the collective evidence for an association with neuroblastoma appears to be supportive but far from conclusive. The study by Daniels and colleagues in this issue 2 is only the second in the literature that has relied on specific information about pesticide exposure 21 rather than on indirect evidence based on paternal job title, family residence or pesticide purchase records. The study was as strong as any, but relies on random digit dialing. The information about children’s exposure to pesticides was elicited from the best possible source, the parents. The results appear to support the hypothesis that pesticide exposure increases the risk of neuroblastoma, but there are some important concerns. It is disquieting that more parental pairs disagreed about exposure to pesticides than agreed that there was indeed such an exposure; this was the case even with respect to extermination, which should be a memorable event. Moreover, information bias cannot be easily discounted in this instance, since many view pesticides with suspicion. Third, pesticides are a large and heterogeneous group, making it difficult to draw generalizable inferences.

Both studies do not provide compelling evidence for a causal link between the studied exposures and outcomes, but they do not provide much comfort either. They are important because they convey two essential messages: (1) the population rates of childhood tumors attributable to these exposures are unlikely to be high and may even be zero and (2) in this particular field, it is difficult to envisage studies more informative than these, unless susceptibility to the exposures under consideration is differentially increased or even limited to particular polymorphisms that need to be evaluated simultaneously.


1. Infante-Rivard C, Olson E, Jacques L, Ayotte P. Drinking water contaminants and childhood leukemia. Epidemiology 2000; 12: 13–19.
2. Daniels JL, Olshan AF, Teschke K, Hertz-Picciotto I, Savitz DA, Blatt J, Bondy ML, Neglia JP, Pollock BH, Cohn SL, Look AT, Seeger RC, Castleberry RP. Residential pesticide exposure and neuroblastoma. Epidemiology 2000; 12: 20–27.
3. La Vecchia C, Levi F, Lucchini F, Lagiou P, Trichopoulos D, Negri E. Trends in childhood cancer mortality as indicators of the quality of medical care in the developed world. Cancer 1998; 83: 2223–2227
4. Wang PP, Haines CS. Childhood and adolescent leukaemia in a North American population. Int J Epidemiol 1995; 24: 1100–1109.
5. Crist WM, Pui C-H. The leukemias. In: Nelson Textbook of Pediatrics. 15th Edition. Eds: Behrman RE, Kliegman RM, Arvin AM. Saunders Publications. 1996:1452–1455.
6. U.S. National Academy of Sciences Committee on Biological Effects of Ionizing Radiation. Washington, DC: BEIR-V (U.S. N A.S.) 1990.
7. Petridou E, Trichopoulos D, Dessypris N, Flytzani V, Haidas S, Kalmanti M, Koliouskas D, Kosmidis H, Piperopoulou F, Tzortzatou F. Infant leukaemia after in utero exposure to radiation from Chernobyl. Nature 1996; 382: 352–353.
8. Linet MS, Hatch EE, Kleinerman RA, Robison LL, Kaune WT, Friedman DR, Severson RK, Haines CM, Hartsock CT, Niwa S, Wacholder S, Tarone RE. Residential exposure to magnetic fields and acute lymphoblastic leukemia in children. N Engl J Med 1997; 337: 1–7.
9. UK Childhood Cancer Study Investigators. Exposure to power-frequency magnetic fields and the risk of childhood cancer. Lancet 1999; 354: 1925–1931.
10. Repacholi MH, Albhom A. Link between electromagnetic fields and childhood cancer unresolved. Lancet 1999; 354: 1918–1919.
11. National Research Council. Possible health effects of exposure to residential electric and magnetic fields. National Research Council, Washington: DC: National Academy Press, 1996.
12. Kinlen LJ. Evidence for an infective cause of childhood leukaemia: comparison of a Scottish new town with nuclear reprocessing sites in Britain. Lancet 1988; II: 1323–1327.
13. MacMahon B. Is acute lymphoblastic leukemia in children virus-related? Am J Epidemiol 1992; 136: 916–924.
14. Petridou E, Kassimos D, Kalmanti M, Kosmidou H, Haidas S, Flytzani V, Tong D, Trichopoulos D. Age of exposure to infections and childhood leukemia risk. BMJ 1993; 307: 774.
15. Greaves MF. Aetiology of acute leukemia. Lancet 1997; 349: 344–349.
16. Foster AM, Prentice AG, Copplestone JA, Cartwright RA, Ricketts C. The distribution of leukaemia in association with domestic water quality in south west England. Eur J Cancer Prevention 1997; 6: 11–19.
17. Kristensen P, Andersen A, Irgens LM, Bye AS, Sundhem L. Cancer in offspring of parents engaged in agricultural activities in Norway: Incidence and risk factors in the farm environment. Int J Cancer 1996; 65: 39–50.
18. Bunin GR, Ward E, Kramer S, Rhee CA, Meadows AT. Neuroblastoma and parental occupation. Am J Epidemiol 1990; 131: 776–780.
19. Spitz MR, Johnson CC. Neuroblastoma and parental occupation. Am J Epidemiol 1985; 21: 924–929.
20. Wilkins JR, Hundley VD. Paternal occupational exposure to electromagnetic fields and neuroblastoma in offspring. Am J Epidemiol 1990; 131: 995–1107.
21. Michaelis J, Haaf HG, Zollner J, Kaatsch P, Krummenauer F, Berthold F. Case control study of neuroblastoma in West-Germany after the Chernobyl accident. Klin Padiatr 1996; 208: 172–178.
© 2001 Lippincott Williams & Wilkins, Inc.