The age distribution of cancer and a multi-stage theory of carcinogenesis

Abstract

ATTEMPTS to derive theoretical laws from changes in the death rate with age have a long history. They have not, in general, been very fruitful and there has been some hesitation in applying the technique to the study of cancer. Recently, however, two hypotheses about the mechanism of carcinogenesis have been put forward, which have been derived from analysis of cancer mortality statistics. Fisher and Hollomon (1951) used statistics from the United States for cancer of the stomach in women, and Nordling (1953) , classing all sites together, used statistics for cancer in men from Britain, France, Norway and the USA. Both found that, within the age group 25–74 years, the logarithm of the death rate increased in direct proportion to the logarithm of the age, but about six times as rapidly; in other words, the death rate increased proportionally with the sixth power of the age. Death rates in some age groups under 25 years were higher than would be expected had this basis been a general law throughout life. Rates for the age groups above 75 years were considered unreliable and were excluded.