When exposures to sparsely ionising radiation such as X-or y-rays are protracted in time, generally the risk of radiationinduced cancer decreases. Qualitatively, a similar result has been observed for neoplastic transformation when the exposure of cells in vitro was protracted (Han et al., 1980;Hill et al., 1984aHill et al., , 1987. However, during the course of our studies in vitro with a densely ionising radiation, unexpected results were obtained: dose protraction led to enhanced transformation. To explain these results a cell-based model was developed (Elkind, 1991(Elkind, , 1992(Elkind, , 1994; a model that also explains enhanced tumorigenesis when doses of high-linear energy transfer (LET) radiation are protracted in time, as in the instance of the inverse relationship between lung cancer risk among uranium miners and radon concentration (e.g. Lubin et al., 1994). The formal properties of this model lead to the inference that, under certain circumstances, the protraction of doses of X-or y-rays could also increase rather than decrease the risk of cancer. In this editorial the radiobiology of the model and its supporting data are described as a basis for applying the model to the question of the increasing incidence of breast cancer.
Neoplastic transformation in vitroIn the 1970s several groups initiated studies of radiationinduced neoplastic transformation in vitro. The end point that was used was focus formation of C3H mouse embryo lOT1/2, Balb/c 3T3 or Swiss 3T3 fibroblasts. These cells express postconfluence inhibition of cell division. Effective cell-to-cell communication results in the down-regulation of growth but not in transformed cells. Thus, densely staining discrete foci of piled-up cells, which had grown in the midst of a confluent layer of normal cells, were readily identified and scored.Because of a long-term interest in the role of repair processes in radiation-induced changes in mammalian cells, early in our work with IOT1/2 cells at the US Argonne National Laboratory we examined the role of repair of sublethal damage as well as damage subeffective in neoplastic transformation. With X-and y-rays, i.e. low linear energy transfer (low-LET) radiations, the effect of repair was examined by applying dose fractionation and/or the protraction of exposures in time via continuous, low doserate irradiation. Qualitatively, our results were consistent with tumour induction in animals. Even though sublethal damage (Elkind and Sutton, 1959) was effectively repaired in lOT1/2 cells (Han and Elkind, 1979), the concomitant repair of subeffective transformation damage resulted in a net