Natural and radiation carcinogenesis in man II. Natural leukaemogenesis: initiation

Publisher Website
Google Scholar
Abstract
The mathematical model of the initiation-phase of human carcinogenesis derived in part I is applied here to the interpretation of ‘spontaneous’ or natural leukaemogenesis. Extensive use is made of mortality statistics and of familial, genetic, cytogenetic and epidemiological studies. It is found that the germ-cell mutation-frequency for 'leukaemogenic’ genes (expressed in terms of: leukaemogenic mutations, per gene at risk, per cell at risk, per year) is, within the limits of computational error, also applicable to somatic cells throughout the entire life span. This value approximates to the upper limit of the normal range of germ-cell mutation-frequencies observed for dominant autosomal and sex-linked loci in man. The initiation of prezygotic leukaemias in adults appears generally to require a cell with one inherited specific gene—or chromosomal-mutation, and three specific somatic mutations. In postzygotic leukaemias—increasing from about 1 % of the natural incidence at 20 years, to about 7 % at 70 years—all four leukaemogenic mutations should be accumulated in one (or more) somatic cells. In chronic lymphatic leukaemias (pre-, or post-zygotic), cells with the appropriate three leukaemogenic mutations probably exhibit selective proliferation; this should be characteristic of many malignancies in adults, especially those of epithelial tissue. When premalignant selective proliferation is present, multi-focal tumour origin may be anticipated. The initiation of the majority of leukaemias in childhood and adolescence probably requires at least one cell with one inherited mutation and two somatic mutations: a condition of ‘stress’ is then necessary for malignant development. This ‘stress’ may be provided by a viral or bacterial infection (for which there is independent evidence). Occasionally, but especially with parental consanguinity, recessive inheritance of childhood malignancies is observed and the aetiology in such cases involves a cell with two inherited (homologous) mutations and two specific somatic mutations; no stress factor is then necessary to complete the initiation-phase. The theory accounts for many qualitative and quantitative details of the familial and epidemiological evidence, and it illustrates the advantages of a mathematical approach.