Abstract
United Nations Scientific Committee on the Effects of Atomic Radiation New York: United Nations (2001)
The UNSCEAR 2001 Report presents a reassessment of the risks to offspring of hereditary disease following parental exposure to radiation. Advances in our understanding of the molecular basis of naturally occurring genetic diseases have increased rapidly with the development of the Human Genome Project and the Report reviews current knowledge on the molecular nature of the genome and the types of mutation that contribute to Mendelian diseases. Spontaneous mutation rates for human genes are discussed in some detail with particular emphasis being placed on sex differences and the effect of paternal age. This is particularly relevant as the Committee has reverted to the doubling dose methodology for risk estimation used in the 1972 BEIR Report with the doubling dose being derived using human spontaneous mutation rates and mouse induced rates. Previous reports have used mouse data for both spontaneous and induced rates but it is argued that experiments using short lived mice, in which sex differences will not be great, are unlikely to provide a reliable estimate of average spontaneous rates in humans. Direct data on spontaneous rates derived from a heterogeneous population of humans will automatically incorporate any sex differences and age effects. This change in conceptual approach is also supported by recent recognition of the role of mosiacs in the development of clusters of mutants in mice and the resulting uncertainties this has introduced into the calculation of mouse spontaneous mutation rates.
A comprehensive account of multifactorial diseases examines the models used to characterise the genetic–environmental interactions which form the etiological basis of this heterogeneous group of disorders. Congenital abnormalities are usually placed into this category since, although upwards of 50% are judged to be attributable to nongenetic causes, the vast majority of the remainder have a multifactorial origin. Similarly, a range of common adult diseases are due to a complex interaction of genetic and environmental factors and the Report examines in some detail current knowledge of genetic susceptibility and risk factors associated with diabetes, coronary heart disease and essential hypertension.
In deriving risk estimates the aim is to predict the impact of a small dose of radiation on mutation rates and hence disease incidence in the population. In order to assess the responsiveness of genetic diseases to induced mutation the concept of the mutation component is applied, this being defined as the relative change in disease incidence per unit relative change in mutation rate. The introduction of this concept has been particularly valuable for estimating the impact of a change in mutation rate on multifactorial diseases over several generations. This has been achieved by using models of disease liability which incorporate an environmental and genetic contribution and introduce mutation and selection as additional parameters. The conclusion from this work is that the mutation component for multifactorial diseases in the first few generations following a permanent small increase in mutation rate is less than 2'. The adoption of this approach has allowed the Committee, for the first time, to derive risk estimates for multifactorial diseases.
The Report also devotes a section to familial cancer and describes a subset of genes in which germinal mutations can subsequently predispose the carriers to cancer. In addition to the 30 or so mutant genes, generally of high penetrance, associated with defined phenotypes it has been postulated that there may also be a much greater number of genes in which mutations cause a lower degree of cancer predisposition. These are difficult to detect and more data is required in order to assess the contribution of germinal mutations to the background load of cancers in the population. Nevertheless, the premise that individuals carrying such mutant genes are also more susceptible to radiation-induced cancer is explored and it is concluded that the presence of such susceptible genomes will have an impact on somatic radiation cancer risks.
The Committee has also reviewed new information from human and animal studies that has been published since the UNSCEAR 1993 Report. It is noted that no significant increases have been found for a range of reproductive outcomes in human studies of the offspring of those receiving childhood radiation therapy, nor have there been any demonstrable genetic effects resulting from the Chernobyl accident. Data on minisatellite mutations from human and mouse studies are described but, since these are not in protein-coding genes, it is considered that their relevance for risk estimation cannot readily be discerned. Analysis of mutations recovered from large-scale mouse experiments is providing evidence for the induction of large deletions non-randomly distributed across the genotype and gives additional insight into the mechanisms involved in radiation-induced mutation induction and the nature of the resultant phenotypes.
Having reviewed the advances in human genetics and current thinking on the nature of radiation-induced mutations, the concepts, data and analyses used for the estimation of genetic risks are presented. Incidence estimates of genetic diseases constitute an integral part of the risk equation. Estimates for chromosomal and multifactorial diseases remain the same as in the 1993 Report but there has been an upward revision for each of the three classes of Mendelian disease, namely autosomal dominant, autosomal recessive and sex linked. Despite a conceptual change in the derivation of the doubling dose, the recommended dose of 1 Gy is the same as that used in previous risk estimations. In recognition of increased knowledge of the mechanisms involved in radiation-induced mutation induction, a potential recoverability correction factor has been introduced into risk calculations to bridge the gap between mutations recovered in mouse experiments and the potential risk of radiation-inducible genetic disease in humans. Human genes are considered with reference to such criteria as gene size, location, normal function and spectrum of spontaneous mutations. This allows the estimation of the fraction of genes that may be responsive to recoverable mutations.
A further departure from previous reports is the recognition that radiation-induced genetic damage results predominantly in multilocus deletions and is therefore more likely to be manifest as multisystem developmental abnormalities rather than single gene disorders. Insights into potential phenotypes come from increasing awareness of naturally occurring microdeletions and their clinical manifestations and include mental retardation, dysmorphic features, growth retardation and serious malformations. For risk estimation of developmental malformations the Committee depart from the doubling dose approach and use data from mouse experiments on radiation-induced skeletal malformations, cataracts and congenital abnormalities diagnosed in utero to derive an estimate of 1,000 per million progeny per Gy for chronic irradiation of either sex and thus 2,000 per million per Gy for irradiation of both parents.
The risks for all other classes of disease are predicted using the doubling dose method and for the first generation following population exposure these are 750 to 1,500 cases for autosomal and X-linked diseases and 250 to 1,200 cases for multifactorial diseases per million progeny per Gy. The risk for autosomal recessive diseases is zero. No separate estimate is given for chromosomal abnormalities, since these effects are assumed to be incorporated within the risks for either autosomal dominant and sex-linked diseases or congenital abnormalities. Overall the predicted risks for the first generation of 3,000–4,700 cases per million progeny per Gy of parental irradiation represent about 0.41 to 0.64' of the baseline frequency of 738,000 per million. The report does not advocate the use of the direct method of risk estimation, noting that no new data of relevance for the application of this method have become available since the 1993 Report. Nevertheless it is recognised that the direct method is useful and that the risk estimate obtained by this method in the 1993 Report is of the same order of magnitude as that given now.
The UNSCEAR 2001 Report is a comprehensive document which not only provides a detailed description of the concepts, parameters and assumptions used in evaluating the impact of radiation exposure on the incidence of heritable disorders, but is also a useful reference source for anyone wanting an up to date review of current knowledge of the nature of genetic mutations associated with human disease.