The aetiology of cancer in the very young.

Epidemiological studies of cancer in young children have implicated a number of environmental factors, which need to be studied in more detail, but it is probably fair to say that the main benefit of these studies has come from the negative findings, which have served to exclude (or at least place an upper limit on the role of) potential risk factors. Our inability to identify environmental causes could mean either that the environment does not substantially affect cancer incidence in young children, or that we are simply not looking in the right places. Most attention has naturally been focused on the known and suspected environmental carcinogens and mutagens. Based on the data summarised in this paper, one possibility is that the most important mutagen is endogenous: 5-methyl-cytosine. If so, factors increasing cancer risk could be those which increase the rate of spontaneous deamination, or impair the efficiency of the excision repair enzymes, or regulate the processes of CpG methylation and demethylation. These factors could still be classed as mutagens in their own right, since they would lead to an increase in uncorrected point mutations, but they would be distinctive in a number of ways. Firstly, since the role of methylation in bacteria is very different from that in eukaryotic cells the Ames mutagenicity assay could not be relied upon to detect methylation-mediated mutagens. Furthermore the risk may be highly age dependent, reflecting changes in methylation patterns with growth and cellular differentiation. Agents which disrupted the imprinting process in the testis would not be detectable by animal carcinogenicity tests unless specifically looked for.(ABSTRACT TRUNCATED AT 250 WORDS)