Relationship between Animal and Stem Cell Dose-Survival Curves

A model of acute mammalian hemopoietic radiation mortality is proposed which relates the dose-survival curve of a population of animals with that of its stem cells. An animal is assumed to survive if it has a certain threshold number of effective mature cells, given by a threshold function, at a specified time after irradiation. A random variable, the "effective proliferation factor," is specified, which describes the proliferation initiated per stem cell. With the assumption that stem cell survival obeys Poisson statistics, the probability that an animal will survive a dose, D, is then uniquely determined. This probability is found to give an approximately normal, S-shaped dose-survival curve, the mean dose and "width" of which are known functions of six stem-cell parameters and five additional parameters. The width of the curve depends principally on: (1) interanimal variation of radioresistance (in turn dependent on the interanimal variations of five parameters); and (2) randomness in the number of stem cells surviving, and in the effective proliferation per stem cell. A principal conclusion is that the steepness of the animal dose-survival curve sets an upper limit on the possible values of the interanimal variation of the stem-cell inverse slope parameter, D₀. The theory is applied to the mouse as a numerical example.