A report by a team of epidemiologists headed by Gerald Draper of Oxford University and published in the British Medical Journal, confirms and updates the findings of the 1984 panel chaired by Sir Douglas Black, which concluded that there was indeed a 'cluster' of childhood leukaemia cases and non-Hodgkin's lymphoma in the village. And it provides new evidence that the cluster, based on statistics for 1963 to 1983, was 'highly unlikely' to have happened by chance.
The Draper analysis shows that in the six years after the Black report, the incidence of the diseases in the 0 to 24 age group remained higher than the national rate and those for the surrounding areas. But the study sheds no light on the cause. In particular, it neither supports nor disproves the controversial hypothesis, put forward in 1990 by Professor Martin Gardner of Southampton University, that there was a statistically significant relationship between a father's radiation dose in the six months before the conception of his child and the risk of the child subsequently developing leukaemia.
Professor Gardner's results have not been confirmed by other studies. For example, a survey of nuclear facilities in Ontario, Canada, published last year, found no increased risk of leukaemia in the children of fathers working at the plants. The Gardner results are further confounded by the fact that the cases were confined to Seascale, even though most of the workers at the Sellafield reprocessing plant live elsewhere.
These and other complications have led a number of scientists to suspect that other factors may be playing an equal, or even greater role. And although most epidemiologists believe that radiation and environmental pollution can be blamed for only a small fraction of childhood leukaemia, the emerging evidence suggests that Sellafield and other nuclear plants may be special cases, albeit with the potential to provide more general insights into the causes of the disease.
Sir Richard Doll, of Oxford University, says: 'It does look as if there's some suggestion that, within the areas around many nuclear installations, there has been an excess of childhood leukaemia.
'But in this country and in Germany there has also been an excess reported in areas where nuclear stations were planned but never actually built. There is a lot to suggest that it is something characteristic of the area, rather than of the nuclear installation.'
Two British scientists - Melvyn Greaves, director of the Leukaemia Research Fund Centre at the Institute for Cancer
Research in London, and Leo Kinlen, director of the Cancer Epidemiology Research Group at Oxford - have set the leukaemia research world abuzz with a pair of related hypotheses, which trace childhood leukaemia to rare or unusual responses to infections early in life, including possible exposure to viruses.
Dr Greaves, in particular, has noted puzzling associations between a child's socio-economic status and the risk of falling victim to leukaemia. Unlike most other diseases, leukaemia poses a four times higher risk to children in developed countries than to those in the developing world.
This correlation was reinforced by a study released last year by the Office of Population Censuses and Surveys. A survey of 10,000 cases of childhood leukaemia and non-Hodgkin's lymphoma, led by Dr Draper's Childhood Cancer Research Group, concluded that the incidence in the wealthiest fifth of the British population was 10 to 15 per cent higher than in the poorest fifth.
Dr Greaves has suggested that the most frequently diagnosed type of childhood leukaemia - common acute lymphoblastic leukaemia - is caused by at least two spontaneous mutations in the child's white blood cells, which are important for the immune response. This model is consistent with the mechanisms now thought to cause other, better understood, types of cancer.
The first mutation occurs while the foetus is in utero, when the cells are immature and rapidly dividing, a situation that makes their DNA more vulnerable to the copying errors that can lead to genetic alterations. The second mutation is triggered later in childhood, when an infection stimulates the mature white cells to divide again.
Dr Greaves proposes that children exposed to viruses relatively late in childhood - including the offspring of wealthier families, who may be more sheltered from other children - may be at higher risk of a second mutation, either because their delayed exposure to infection results in a more powerful immune response or through a direct effect of the virus on their bone marrow where the white cells are manufactured.
But Professor Kinlen suspects that viruses may be a more direct cause of certain cases of childhood leukaemia, analogous to the role of HTLV-1 virus in some types of adult leukaemia. Such 'transforming' viruses cause genetic damage when their genetic material becomes incorporated within the DNA of the target cell.
Professor Kinlen bases this view on a number of epidemiological studies he and his colleagues have performed on the effects of population mixing. The most dramatic finding concerns the new towns constructed in post-war Britain to relieve overcrowding in large cities; those built in rural areas had leukaemia rates up to six times greater than expected.
These results, Professor Kinlen argues, are most consistent with a viral explanation. 'When many people come together from different geographic regions,' he says, 'an increase of contacts occurs between infected and susceptible individuals: the basis for the spread of any infection.'
Sir Richard Doll finds the Greaves hypothesis 'biologically very attractive', because it would explain the increase in childhood leukaemia in Britain that occurred between about 1920 and 1945, during a period when standards of health and hygiene dramatically improved.
The ideas of Dr Greaves and Professor Kinlen may eventually help provide a better understanding of the mechanisms behind childhood leukaemia, as well as a clearer explanation for the anomalies of the situation at Sellafield.
'It may be that radiation-induced damage to the sperm of the fathers at Sellafield initiated a predisposition to leukaemia,' Dr Greaves says, 'but that it was insufficient in itself. Something else may have been required - the sort of virological effect that Kinlen and I have been proposing.'