Page 74 - 20dynamics of cancer

4 History of Theories

In this chapter, I discuss the history of theories of cancer incidence. I
focus only on those aspects of history that remain relevant for current
research on progression dynamics and incidence. More details about the
history and the literature can be obtained from the many published ar-
ticles that review theories of cancer incidence (Armitage and Doll 1961;
Druckrey 1967; Ashley 1969b; Cook et al. 1969; Doll 1971; Nowell 1976;
Peto 1977; Cairns 1978; Whittemore and Keller 1978; Scherer and Em-
melot 1979; Moolgavkar and Knudson 1981; Forbes and Gibberd 1984;
Stein 1991; Tan 1991; Knudson 1993, 2001; Lawley 1994; Iversen 1995;
Klein 1998; Michor et al. 2004; Moolgavkar 2004; Beckman and Loeb
2005).
The first section introduces the original theories of multistage pro-
gression. Starting in the 1920s, several experimental programs applied
chemical carcinogens to animals. Two different carcinogens applied in
sequence often yielded a higher rate of cancer than did application of
a single carcinogen over the same time period. This synergistic effect
between two carcinogens led to the idea that each carcinogen stimulated
a different stage in progression: the two-stage model of carcinogenesis.
A separate line of multistage theories began in the 1950s by analysis
of the observed rates of cancer at different ages. For most of the com-
mon adult cancers, the age-specific incidence curves rise with a high
power of age, roughly proportional to t n−1 , where t is age and, from the
data, n ≈ 6. Mathematical models showed that such incidence curves
would occur if cancer follows after progression through n rate-limiting
steps. This analysis led to the hypothesis of multistage progression.
The second section turns to the most profound empirical tests of
multistage theory. The mathematical theory predicted that the greater
the number of rate-limiting steps, n, the faster incidence rises with age.
Ashley (1969a) and Knudson (1971) reasoned that if somatic mutation is
the normal cause of progression, then individuals who inherited a muta-
tion would have one less step to pass before cancer develops. Multistage
theory makes the following prediction: inherited cases with a smaller