Since the Chernobyl accident in 1986, a number of epidemiological studies have been published about rates of leukaemia in children in various countries and the possible link between these rates and the radioactive contamination caused by the accident. The committee focused on infant leukaemia (ie leukaemia occurring in the first year of life). It is believed that the biological origin of infant leukaemia is different to that of childhood leukaemia (ie leukaemia occurring after the age of one), and that infant leukaemia may be linked to mothers' intakes of substances that affect DNA. The reason for the committee's focus on this topic is that it has been asserted that increases in infant leukaemia after Chernobyl provide unequivocal evidence that current risk estimates for internal radiation emitters are in error, perhaps by a factor of more than 100.

There are six countries in which increases in rates of infant leukaemia have been reported in the literature: Scotland, Greece, western Germany, the US, Wales and Belarus. There is also a relevant study for Europe as a whole that examined childhood leukaemia, but in which rates specifically for infant leukaemia are presented. The position on these studies is as follows.

Scotland

The published study shows that there were 9 cases of infant leukaemia registered in Scotland in 1987 and the first half of 1988, which is over four times more than would be expected. The geographical distribution of these cases is not given in the study, so it is not possible to say whether they occurred in areas where contamination was highest. The committee agreed to seek the geographical information from the authors. Meanwhile, it noted that whole body scans of some of the infants and their mothers had not detected any increased levels of caesium-137.

Greece

This study shows a 2.6 fold increase in infant leukaemia cases in children born in the period July 1986 to December 1987. There were 12 cases in Greece in this 18 month period. There was a statistically significant trend for there to be more cases in areas of higher contamination.

Germany

There were about one and a half times more infant leukaemia cases in western Germany among those born between July 1986 and December 1987 than would be expected. The authors of the study analysed the geographical distribution of these cases and found that the highest numbers were in the areas with the lowest levels of contamination.

United States

A study of registrations in 12 US states and cities found that there were 62 cases of infant leukaemia among those born in 1986 and 1987, which is 30% more than in the previous five years and the subsequent two years. This increase is not statistically significant and levels of contamination in the US following the Chernobyl accident were much lower than those in Europe.

Wales

The 4 cases of infant leukaemia in Wales registered in 1987 and 1988 are an increase over the number expected. A query about whether 3 of these cases occurred in the first or the second half of 1988 is being pursued by the committee and it is also seeking information about the geographical distribution of these cases.

Belarus

The study of infant leukaemia found about one and a half times more infant leukaemia cases born in the period from July 1986 to December 1987 in two areas of Belarus where deposition levels were highest, and a rate ratio of 1.26 for Belarus as a whole. Neither of these increases was statistically significant. Neither an earlier nor a more recent, follow-up study found a detectable increase of childhood leukaemia overall in Belarus. The committee has requested a more detailed breakdown of the data from this follow-up study.

Europe

In a study across much of Europe, the International Agency for Research on Cancer (IARC) did not detect a raised risk of childhood leukaemia overall after the Chernobyl accident. The committee has asked for a more detailed breakdown of the IARC data and is also enquiring about a study specifically of infant leukaemia that is in progress.

The committee concluded that the question of infant leukaemias following the Chernobyl accident warrants further investigation. It will return to the topic when it has received responses to its various enquiries. At that time three questions will be addressed:

1. Could the increases in infant leukaemias in the various countries be a chance effect?
2. Was it irradiation of the embryo/foetus from Chernobyl? Is there any other possibility?
3. What does this tell us about current infant leukaemia risk factors?

Advances in molecular genetics over the last few years have enabled new techniques to be employed to examine alterations in DNA brought about by irradiation. Since the late 1990s a number of studies have been published about mutations in minisatellite DNA sequences in children born to parents who have been exposed to radiation. These minisatellites are sequences of repeated elements of ten to a hundred nucleotides in length. They are not part of the coding message in the human genome but may affect transcription of adjacent genes. Taken as a whole, the published studies show significant increases in minisatellite mutations following intakes of radionuclides by the parents.

At present, there is not enough information to relate the minisatellite mutation rates to radiation exposure in a quantitative way. There are also difficulties in determining the consequences of the minisatellite mutations for the health of the children and their offspring. These difficulties arise because minisatellite mutations occur at fairly high rates in the absence of radiation exposure and seem to be linked to diseases not normally associated with radiation.

A further paper on minisatellite mutations is expected to be published shortly, at which time the committee will return to the question of the significance of minisatellite mutations for health risks. The committee is also contacting the authors of the study that shows the highest rates of minisatellite mutations, to ask a question about validation of their methods.


The committee discussed quantitative comparisons of cancer induction in humans by internal and external irradiation. Many of the data that are able to be compared quantitatively show that risk factors for internal irradiation are similar to those for external irradiation (within a factor of 2-3, upwards and downwards). For leukaemia in populations of mixed ages the risk per unit dose for external irradiation may be a factor of about ten higher than that for internal irradiation by some alpha emitting radionuclides.

The estimates of dose and risk from intakes of radionuclides depend on assumptions about the distribution of the radionuclides within body organs and tissues and about target cells for cancer induction. The committee will be dealing with these matters in future meetings. There are uncertainties in risk estimates for intakes of radionuclides that arise from uncertainties about people's radionuclide intakes and from the quality of cancer incidence and mortality data. There are also questions about differences between natural and artificial radionuclides, which will be discussed at future meetings.

Next (Meeting 3)