Reasons why the idea that radiation exposures induce cancer needs to be revisited

It has long been thought that the carcinogenic effect of radiation resulted from the induction of oncogenic mutations which then led to an increase in the proportion of cancer-bearing individuals. However, even as early as the 1960s, there were indications that the carcinogenic effect of radiation might result from the induction of an earlier onset of cancer. Recently, the former notion was challenged by its inability to explain time-dependent decline of the relative risk following an exposure to radiation, and a parallel shift of mouse survival curves toward younger ages following an exposure to radiation. The two observations are clearly understood if it is assumed only that a radiation exposure causes an earlier onset of spontaneously occurring cancers.

In the present study, a critical review was conducted which examined papers that showed dose responses which apparently supported the mutation induction theory of radiation carcinogenesis.

It was found that there were two types of misleading experimental designs: one consisted of studies in which observations were prematurely terminated, and which consequently hid a complete story of radiation carcinogenesis. The other set of papers used age adjustments which were derived from the idea that the life shortening effect of radiation needs to be compensated for since tumor mortality becomes higher among older subjects. This type of adjustment appeared reasonable but was found actually to be a different form of description on an earlier onset of cancer following radiation exposures.

In mouse experiments, radiation exposures did not lead to the induction of a large increase in the proportion of tumor deaths when life-long observations were made. Human epidemiologic data are also in line with the earlier onset hypothesis of radiation action. It should be cautioned, however, that the earlier onset model applies only to malignancies whose mortality increases rapidly with the increase of age and does not apply to diseases of short latency such as childhood leukemia and thyroid cancers.