Henry S. Kaplan Distinguished Scientist Award 2003. The crooked shall be made straight; dose-response relationships for carcinogenesis.

Estimates of radiation-induced malignancies come principally from the atomic (A)-bomb survivors and show an excess incidence of carcinomas that is linearly related to dose from about 5 cGy to 2.5 Gy. Above and below this dose range there is considerable uncertainty about the shape of the dose-response relationship. Both the International Commission of Radiation Protected (ICRP) and the National Council of Radiation Protection (NCRP) suggest that cancer risks at doses lower than those at which direct epidemiological observations are possible should be obtained by a linear extrapolation from higher doses. The demonstrated bystander effect for irradiation exaggerates the consequences of small doses of radiation and implies that a linear extrapolation from high doses would underestimate low dose risks. It is possible to make estimates of the cancer risk of diagnostic radiological procedures. Helical computed tomography in children is of particular interest since it is rapidly increasing in use and the doses involved are close to the lower limit of significance in the A-bomb survivors. For example, an abdominal computed tomographic scan in a 1-year-old child can be estimated to result in a lifetime cancer risk of about 1:1000. In the context of radiotherapy, some normal tissues receive 70 Gy, while a larger volume receives a lower dose, but still far higher than the range for which data are available from the A-bomb survivors. Data are available for the risk of radiation-induced malignancies for patients who received radiotherapy, e.g. for prostate or cervical cancer. New technologies such as intensity modulated radiation therapy could result in a doubling of radiation-induced second cancers since the technique involves a larger total-body dose due to leakage radiation and the dose distribution obtained involves a larger volume of normal tissue exposed to lower radiation doses.