While the diseases listed in Table 5.1 presumably develop as a result of mutations within a single gene, mostly unique to that disease, this knowledge tells us little about how the disease actually develops. In fact, if the mutation is in the germline, then every cell of the organism possesses a copy of that mutated gene within its nucleus. If such a mutational change were truly dominant in all cells, the host would be expected to develop neoplasms of numerous if not all tissues. This does not occur, although in many diseases listed in Table 5.1, both dominant and recessive, neo- plasia of several tissues may develop. Thus, the expression of the abnormal gene that leads to the neoplastic change must require a number of other accompanying changes that may be genetic, environmental, or both.
A reasonable solution to this dilemma was first proposed by Knudson (1971), who hypoth- esized a two-mutational (two-hit) theory of carcinogenesis. His theory was developed primarily to explain the epidemiological findings seen in hereditary retinoblastoma, in which nearly two- thirds of the hereditary cases were bilateral but all of the sporadic cases were unilateral. The former cases occurred in very young patients (mean age, 18 months), whereas the sporadic uni- lateral cases developed at an average of 30 months or more. On the assumption that mutational events occur at random and a relatively fixed rate, Knudson reasoned that at least two mutational events, now understood to be frequently in each allele of the same gene, were necessary to con- vert a normal cell into a neoplastic cell. If the first mutation or hit were postzygotic (spontaneous or sporadic), the progeny of this mutated cell would then be at an increased risk of developing into a neoplasm when one or more cells received a second hit involving the normal allele. How-ever, since spontaneous mutagenic events are of the order of 1 in 105 per genetic locus (Bridges et al., 1994; Glickman et al., 1994) the chance of a spontaneous two-mutational event would be 1 in 1010, which is highly unlikely and conforms, at the cellular level, to a very rare occurrence, as is generally true of spontaneous cancer. If, however, the first hit were present in the germline (prezygotic), with all cells in the organism having a mutation in one allele of the disease gene, then a second hit would be 105 times more common. Thus, in such individuals, neoplasms would be expected to develop much earlier, more frequently, and bilaterally in tissues in which neo- plasia commonly developed. Furthermore, in this model, the neoplastic susceptibility would be transmitted as an autosomal dominant trait, as noted clinically (Bolande and Vekemans, 1983).
A diagram of the two-hit theory of Knudson is seen in Figure 5.3.
Figure 5.3 Diagrammatic representation of the two-hit theory of Knudson and associates. (Reproduced from Junien, 1989, with permission of author and publisher.)
When first proposed, this theory or model was felt to be potentially applicable to certain germline-dominant diseases associated with a high incidence of neoplasia. Today, Knudson’s theory is regarded as a major basis for our understanding of the process of carcinogenesis, the pathogenesis of neoplasia. Furthermore, the theory set the stage for a more complete understand- ing of many of the genetic and environmental factors in neoplastic development. Thus, while almost all of the evidence points to a genetic basis for the conversion of a normal to a neoplastic cell, Knudson’s theory made a clear distinction between germline genetic changes and somatic (nongermline) genetic changes. In the clinically autosomal dominant conditions leading to neo- plasia, the germline mutation put all cells of the organism at some degree of risk for a somatic mutation that would result in a neoplastic cell. In the development of nonhereditary cancer, both mutations or hits must have occurred in both alleles in individual somatic cells. If the two hits occur, one in each allele of the gene whose abnormal function leads to the development of neo- plasia, then the normal or wild-type gene must give rise to a product that protects the cell from undergoing the neoplastic transformation. Expression of only one normal copy of the gene is sufficient to prevent the appearance of the neoplastic phenotype. The genotypes of normal and neoplastic cells in wild-type, heterozygous, and homozygous abnormal cells may be seen in Table 5.2. This table shows the number of somatic events required before neoplasia develops, along with the genotypes of the host cell and the neoplastic cell. Because of their function in preventing or suppressing the neoplastic transformation of somatic cells, such genes have been termed tumor suppressor genes. A more complete discussion of these genes appears later in this chapter.
Table 5.2 Tumors Caused by Recessive Tumor Suppressor Genes in Hosts of Different Genotypes