Causes of childhood cancer

There need to be a number of genetic mutations within a cell before it becomes cancerous. Sometimes a person is born with one of these mutations already present. This doesn’t mean that they will definitely get cancer, but it makes it more likely. This is called ‘genetic predisposition’.

This genetic predisposition may either be inherited or the result of a genetic mutation which occurs when the child is in the womb:

• Retinoblastoma is an example of a type of cancer which is known to be caused by an inherited faulty gene in some children (two out of five cases are inherited)
• Children born with a genetic predisposition to leukaemia, on the other hand, are known to acquire the predisposing genetic mutation whilst they are in the womb.

The vast majority (90%) of children born with the faulty Rb1 gene develop retinoblastoma. The picture for leukaemia is very different in that for every child who develops leukaemia – around 100 have the mutation but don’t develop the disease.

In most children born with a genetic predisposition, whether inherited or acquired, it appears that a further trigger is required for progression to overt disease.

In the case of childhood leukaemia, the ‘two-hit hypothesis’ proposes that initiating events take place whilst the child is still in the womb. The second ‘hit’ occurs later in life, triggering the development of full-blown leukaemia.

Scientists are trying to establish what form these ‘hits’ take – i.e. what factors (other than spontaneous error) cause the initial genetic mutations and what factors trigger progression of the disease.

Despite a wealth of research, much uncertainty remains over what causes cancer in children.

Many different factors have been linked with the development of childhood cancer, with varying degrees of certainty.

Research is complicated by the fact that there are many different factors which may cause cancer in children. Exposure to more than one of these factors is probably necessary – and probably at different stages of a child’s life.

The relative rarity of childhood cancers further impedes research.

Leukaemia is better represented in research literature than other forms of cancer because it affects more children, making it easier to obtain meaningful results in epidemiological studies. International collaborations are important as they increase the number of cancer cases available for study.

The link between childhood cancer and ionising radiation is well-established. Some of the first evidence came from a study of children whose mothers received abdominal X-rays during pregnancy, a practice which is now avoided.

Further evidence comes from studies of the Japanese atomic bomb survivors. There has been a general excess of cancer in the exposed population, but the greatest excess was for leukaemia. The risk for those exposed at young ages was especially pronounced.

The Chernobyl nuclear reactor disaster in 1986 resulted in a marked excess of thyroid cancer in children in the vicinity of Chernobyl. The excess, which began to appear five years after the incident, is apparent in children who were less than 10 years of age at the time of exposure.

Reports during the 1980s and 90s of excesses of childhood leukaemia in the vicinity of certain nuclear installations (around Sellafield in England, Dounreay in Scotland and La Hague in France) caused a great deal of concern.

Radiation doses from exposure to discharged radioactivity, however, were considered to be too low to account for the cluster. The most favoured explanation for these – and other – clusters relates to unusual patterns of population mixing (see “Infections” below).

It has been suggested that exposure to natural background radiation, to which we are all constantly exposed, may be a factor in the aetiology of childhood cancer, particularly leukaemia.

Although not proven, published risk estimates suggest that 15-20% of childhood leukaemia cases in Great Britain could be attributable to this exposure.

Electric and magnetic fields: There has been much concern about the possible health effects of the electric and magnetic fields associated with both power lines and electrical sources in the home.

Since the first study reporting an association between childhood cancer and electrical wiring was published, at least 25 further studies have been published including three major pooled analyses, all three of which were supportive of a link.

The largest single study in England and Wales to date reported a significantly increased risk of leukaemia in children living within 600 metres of a high voltage overhead power line.

The epidemiological evidence has led to the International Agency for Research on Cancer (IARC) to classify extremely low-frequency electric and magnetic fields (ELF EMF) as ‘possibly carcinogenic’ to humans.

Research aimed at finding a biological mechanism for the health effects of EMF is ongoing.

UV radiation: The ultra-violet component of sunlight is known to increase the risk of skin cancer in adults. Australia and New Zealand have a relatively high incidence of childhood melanoma which may be due to UV radiation.

The possible role of infections in childhood cancer has been much-studied.

Viruses are known to be implicated in some human cancers including:

• Burkitt lymphoma
• Hodgkin lymphoma and nasopharyngeal carcinoma (all associated with Epstein-Barr virus)
• liver carcinoma (hepatitis B) and;
• Kaposi sarcoma (HIV and HHV8).

These associations can only account for a tiny proportion of childhood cancer in western countries.

There is a great deal of support for a role of infection in the development of childhood leukaemia. Acute lymphoblastic leukaemia (ALL) has an incidence peak at 2-4 years of age, coinciding with the timing of common childhood infections such as measles. However, despite intensive research efforts, no specific leukaemia-causing virus has been identified in children.

Two different theories suggest that rather than being the result of a specific leukaemia-causing virus, childhood leukaemia could be the result of an abnormal response to a common infection.

• The ‘delayed infection’ hypothesis suggests that a lack of exposure to infection early in life results in a poorly primed immune system which reacts inappropriately when the child is subsequently exposed to infection.
• The ‘population mixing’ hypothesis also suggests that leukaemia is a rare response to a common infection in susceptible children.

Case study: Sellafield

A cluster of cases of childhood leukaemia around the nuclear reprocessing plant of Sellafield in Cumbria in the 1980s caused speculation that the cluster must be related to radiation exposure.

However, assessments showed that radiation doses were much too low to account for the cluster.

The isolated village of Seascale, at the centre of the cluster, had seen a rapid influx of families. They were brought together from all over the UK to work at Sellafield, and undoubtedly bringing with them an influx of infections.

This provided a basis for suggesting that some childhood leukaemia clusters might be an unusual outcome of a common infection arising in non-immune individuals following ‘population mixing’.

Although extremely difficult to prove definitively, this theory has been borne out by studies of other leukaemia clusters.

Many types of chemicals are known to be carcinogenic. The question is whether children are exposed to high enough levels to lead to the development of cancer. Exposure in utero may also be important.

• Chemicals contained in air pollution – including benzene – are known to be carcinogenic. The levels to which most children are exposed are extremely low and the risks are difficult to detect because exposure is ubiquitous. There is some evidence of an increased risk of childhood cancer for children with birth addresses within 1km of hot spots for various air pollutants
• Associations have also been reported with exposure to pesticides
• The evidence for a link between childhood cancer and parental smoking is mixed. A meta-analysis of more than 30 studies showed a 10% increase in risk of all cancers with maternal smoking during pregnancy, but no evidence for an increased risk of any specific cancer. At least one study has found increased chromosomal abnormalities in amniocyte cells of foetuses of smoking mothers
• Some types of medication and drugs have been linked to childhood cancer. There have been reports of the possible carcinogenic effects of many different drugs taken by mothers during pregnancy. The only one firmly established as a transplacental carcinogen is diethylstilboestrol (DES), a hormone which in some countries used to be given to pregnant women to prevent miscarriage but which has now been discontinued. DES caused an unusual type of cancer in girls and young women
• Some of the drugs used in chemotherapy are known to carry a risk of causing secondary cancer. A major aim of ongoing research is to develop drugs which are less toxic.

It is generally accepted that a healthy diet with plenty of fresh fruit and vegetables can protect against cancer in adults.

At least one study has shown a protective effect in children: Kwan and colleagues demonstrated a strong protective effect against childhood leukaemia risk if oranges and bananas were consumed on a regular basis during the first two years of life.

There is a fairly substantial body of evidence pointing towards a small protective effect of breast-feeding on childhood leukaemia risk.