Intended for healthcare professionals

Editorials

The causes of childhood leukaemia

BMJ 2005; 330 doi: https://doi.org/10.1136/bmj.330.7503.1279 (Published 02 June 2005) Cite this as: BMJ 2005;330:1279
  1. Heather O Dickinson, principal research associate (heather.dickinson{at}ncl.ac.uk)
  1. University of Newcastle upon Tyne, Centre for Health Services Research, Newcastle upon Tyne, NE2 4AA

    Delayed exposure to infection may trigger leukaemia after prenatal damage to DNA

    About one child in 2000 develops leukaemia before the age of 15: most cases are acute lymphoblastic leukaemia.1 Childhood leukaemia is a biologically diverse disease, so several pathways to its development are possible. All probably combine genetic susceptibility and exposure to external risk factors at a time when the child is vulnerable.

    Children with Down's syndrome and certain other genetic syndromes are much more susceptible to leukaemia, but such conditions are involved in relatively few cases.2 Bigger babies3 and boys1 have a slightly higher risk.

    Many external factors have been reported as associated with an increased risk of childhood leukaemia.4 Some may be causal, but some may merely be correlated with the actual cause. Other apparent associations may be due to chance or bias: researchers doing case-control studies often have difficulty selecting controls who are representative of the population. Factors are more likely to be causal if they are biologically plausible, are seen consistently in different populations, and have a dose-response relation to the risk of leukaemia.

    In the 1950s a very large case-control study of childhood cancer in the United Kingdom found that radiography of a mother's abdomen during pregnancy increased her baby's risk of developing childhood leukaemia by about 50%.5 This association is now widely accepted as causal: it was confirmed by meta-analysis of studies in several other populations; babies have higher risks if their mothers have higher doses of radiation; and similar effects were seen in animal experiments.w1 Practically no pregnant women have abdominal radiography nowadays.

    Analysis of high quality data from population based registries in many different countries shows that the spatial and temporal distribution of acute lymphoblastic leukaemia is consistent with the possibility that an unusual pattern of exposure to infection increases the risk.6 Over the past decade, evidence has accumulated to support two hypotheses, which are not mutually exclusive, about the role of infections.

    Kinlen suggested that an influx of population, especially into a previously isolated rural area, could promote epidemics of common infections—probably viruses—to which childhood leukaemia is a rare response.7 In eight major studies in settings with extreme population mixing—new towns, commuter belts, major construction sites, and country areas used for wartime evacuation or for military camps—Kinlen found an increased incidence of childhood leukaemia. Other researchers have confirmed these findings, found that the risk increased as the level of population mixing increased, and estimated that population mixing could be involved in up to half the cases of childhood leukaemia in rural areas.6

    Greaves proposed that common acute lymphoblastic leukaemia (the major subtype) is caused by at least two events: one before birth which is then followed by a challenge from infections after protection from them in early postnatal life.8 Recent molecular evidence has corroborated this theory. About a quarter of children with acute lymphoblastic leukaemia have fusion of TEL and AML1 genes, and these chromosomal translocations have usually developed before birth. But many more babies have these preleukaemic cells in their cord blood than actually develop leukaemia, so postnatal events must be necessary to trigger the disease. Further, epidemiological studies indirectly support the theory that early protection from infection increases a child's risk of leukaemia. Studies have had to use surrogates—such as use of day care—for early exposure to infections. Many have found that babies who go to day care at a young age have a lower risk of developing childhood leukaemia than others. A report by Gilman et al (p 1294) in this issue strengthens this evidence.9 It found a doseresponse relation: babies had greater protection from leukaemia if they had more contacts with other infants.

    Many reports that other factors increase the risk of leukaemia have not been consistently replicated.4 Factors supported by more consistent evidence include being born to a mother over 35 years of age or to one who previously had spontaneous abortions, not being breast fed, having specific haplotypes and genes associated with DNA repair, and certain paternal occupational exposures.2 6 10

    For over 20 years, extremely low frequency magnetic fields—which are produced by alternating electric currents such as those found in high voltage power lines, electric appliances, and wiring—have been suspected of increasing the risk of childhood leukaemia. In 2000 a meta-analysis of nine studies did not find any increase in a child's risk of leukaemia in magnetic fields averaging under about 0.4 μT, but above this level the risk doubled.11 However, the authors suspected that this apparently increased risk could have been due partly to bias in the selection of controls. A magnetic field of 0.4 μT is very weak and amounts to about 1% of the earth's magnetic field, which affects all of us all the time. Studies of animals and of cells in culture have found no evidence of a plausible biological mechanism whereby such very weak magnetic fields could influence the development of leukaemia.w2

    In this issue (p 1290), Draper et al report on a very large case-control study, which found that a child's risk of leukaemia increased steadily with proximity to high voltage power lines of the home they lived in at birth.12 However, this study did not include estimates or measures of the magnetic field from either the power lines or other sources. So it provides little evidence that the increased risk closer to power lines is due to magnetic fields. Furthermore, its matching of controls to cases on the basis of administrative areas may have yielded controls who were not completely representative of the distance of children's homes from power lines. Finally, the risk of childhood leukaemia varies geographically, so the increased risk closer to power lines may reflect some other factor that varies geographically. Even if the effect is causal, it could account for only a tiny proportion of cases.

    We don't yet fully understand the aetiology of childhood leukaemia. Nevertheless, we are now reasonably sure that it often involves damage to DNA before birth—probably in response to infection, chemicals, ionising radiation, or other environmental exposures.6 These preleukaemic cells are converted into overt disease after birth if children are susceptible—because of their genetic make up and early protection from infection—and experience one or more further events, often a delayed challenge from infections. Further insights will almost certainly come as advancing technology helps us to understand the molecular events that drive leukaemic changes.

    Footnotes

    • Embedded Image Additional references w1 and w2 are on bmj.com

      Papers pp 1290,1294

    • Competing interests None declared.

    References

    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    View Abstract

    Log in

    Log in through your institution

    Subscribe

    * For online subscription