Introduction
83The most straightforward way to assess the 87 relative sensitivity of different tissues to the induction of tumours by radiation would be to 89 compare the induced tumour rates when all the 89 tissues received the same dose at the same time. 94 No assumption need then be made about the form of the close-response relation except that it 94 is the same in all tissues. The only close approach to this ideal situation in man is the exposure of the 95 foetus in antenatal radiography. In all the other 56 RADIOSENSITIVITY AND SPATIAL DISTRIBUTION OF DOSE 57 Circumstances in which large fractions of the body, and therefore many different organs, have been irradiated, therapeutically, accidentally, occupationally, or in war, the dose received by different organs has varied considerably, even when the organs have been "directly" irradiated. For example, it has been suggested that the dose in the stomach in an ankylosing spondylitic given X-ray therapy was about one-tenth of the dose in the overlying spinal marrow (Dolphin and Eve, 1968). An allowance for the difference in dose has to be made if the observed tumour incidences in the different organs are to provide information about the relative sensitivity of these organs to tumour induction.The ICRP assumes a linear relation between dose and biological effect. This is certainly the simplest function and will be used in what follows. The effect of choosing a different form of dose-response is considered later.A linear relation between radiation dose and tumour induction has implications for the mechanism of carcinogenesis by radiation. It cannot be expected to hold when the dose is high enough to produce gross tissue damage. When a high proportion of tissue cells is killed, tumour yield may well be reduced and when healing is imperfect, or fibrosis is marked, tumour yield may be increased. Extrapolation from such situations is primafacie a highly unreliable guide to what might be expected at dose limit or emergency levels of exposure. Therefore, information about tumour induction following localized therapeutic irradiation for cancer has not been used in formulating our conclusions. Also excluded is all information about tumour induction by radioactive materials that are selectively concentrated in particular organs--such as the bone-seekers and the isotopes of iodine and phosphorus. The problems of estimating the ratio of dose in the concentrating tissue and in the adjacent tissues are formidable (cf. ICRP Publication 11, A Review of the Radiosensitivity of the Tissues in Bone, 1968).Particularly useful information is available from three categories of human subjects--British patients with ankylosing spondylitis irradiated as adults, Americans irradiated in infancy or childhood for a supposedly enlarged thymus or for a variety of childhood complaints, and Japanese of all ages when exposed to an atom bomb explosion. In each category there was a range of exposure levels, and an approximate dose-response can be ascertained for two kinds of malignant diseas...