Radiation, the two-edged sword: cancer risks at high and low doses.

Diagnostic Radiology and Radiation Oncology represent beneficial uses of radiation. The possible price tag is that radiation can cause, as well as diagnose or cure, cancer. Cancer risks at high doses are well known from epidemiologic studies of the Japanese survivors. Risks at low doses must be extrapolated from the high-dose data. The standard-setting bodies recommend a linear no-threshold extrapolation, but this is controversial. A knowledge of mechanisms, while not replacing epidemiology as a source of radiation-induced cancer risks, may provide insights into the shape of the dose-response relationships, and therefore on the validity of the linear extrapolation. The spectrum of second malignancies in radiotherapy patients sheds some light on mechanisms. More useful are new experiments involving the single-particle microbeam; these have demonstrated a bystander effect, that is, biologic effects in cells that are not directly irradiated, as well as mutations in cells where the cytoplasm, but not the nucleus, has been traversed by an alpha-particle.