Malignancies following low radiation exposures in man

It is very hard to estimate the risk of malignancy entailed in exposure to low doses of ionizing radiation. The risk from higher doses (100 rad or more) can now be estimated approximately from several sources of epidemiological evidence, both for exposure of the whole body and for that of a number of organs or tissues. Many of these sources are imperfect in some respects, often involving inadequate periods of follow-up, deficient control data, uncertainty of dose at the site of interest or of RBE for exposures to radiation of high linear energy transfer (LET), or gross statistical imprecision if only small excess numbers of tumours are observed. Together, however, these data are beginning to justify estimates of the frequencies with which malignancies are induced in man in various tissues by moderately high absorbed doses, often delivered at high dose rate. Until recently there was little information on the frequencies of tumour induction by doses of only a few rad, and no reliable basis for inferring this frequency from that observed at considerably higher doses. Moreover, the great variability in the frequency of tumour induction by radiation in different animal species, and even in different strains of the same species, prevents any quantitative inferences being applied directly to man. Three pieces of evidence now appear to indicate the induction of malignancies in man by absorbed doses of less than 10 rad.