660 Sutow and Conard in the proximally exposed and 6 per million in the distally exposed children.* * Straightline relationships between irradiation doses and leukemia incidence in the Japanese data were noted over the dose range from 100 to 900 rads. The British data were obtained from follow-up studies on about 15,000 patients, predominantly males of 14 years and older, who were given radiotherapy for ankylosing spondylitis‘! ** #1; 73 cases of leukemia have occurred among these patients. The number of cases of leukemia expected in this study population in the absence of irradiation has been estimated to be four or five.'! These patients received therapeutic doses of radiation; therefore, “this study does not provide evidence on leukaemia incidence after doses below 500 rad.’’*? Other epidemiologic investigations have indicated that children with leukemia have had greater exposure to radiotherapy than have comparison groups??-%* and that children subjected to therapeutic irradiation, particularly to the thymic area, have a higher incidence of leukemia than expected.* * 73? In some reported series, however, children who received therapeutic irradiation, including irradiation of the thymus, did not show a significant incidence of leukemia.**-** Again, the history of therapeutic radiation is not always a significant finding in leukemic children.’ *7 The total volume of tissue irradiated (bone marrow) has been suggested as a factor of importance.*° Although increased frequency of deaths from leukemia among American radiologists has been reported,‘ 4? the data do not contribute toward quantitative aspects of the problem.?**¢ Several attempts have been madeto determine the probability of leukemogenesis on the basis of radiation dose.?* *4348 The conclusions do not definitively answer the question of the existence or nonexistence of a threshold radiation dose for the development of leukemia in man.*® A cause and effect relationship, however, between high-dose radiation exposure in man and increased incidence of leukemia must be accepted.*° radiation and the occurrence of malignant neoplasia other than leukemia has been the object of intensive scientific inquiry since the publication of reports of increased carcinogenesis in children who wereirradiated in infancy for thymic enlargement.* > Cancer of the thyroid gland appears to be the most frequent malignant tumor noted in these studies.*-7* #9 Eleven cases of thyroid cancer developed in a group of 1,644 children given x-ray therapy to the head, neck, or chest. In the same population, 0.12 case would have been expected. Irradiation dosage in the 11 children ranged from 100 to 1,770 r in air.® In another prospective study of 2,809 infants who had been therapeutically irradiated for thymic enlargement, 9 cases of thyroid cancer were found (0.10 case expected) in addition to 21 cases of thyroid adenoma (0.9 case expected). The 9 with thyroid cancer had cumulative exposures ranging from 156 to 1,092 r in air with a mean of 598 r.’ Al- though the development of “thyroid nodules” has been ascribed to prior therapy with iodine-131,** *° only one case of thyroid carcinoma has been reported among patients so treated.*® °° Various data suggest that the formation of the nodules was more frequent in children than in adults (33.3 per cent compared to 0.84 per cent).®° Clinical observations have indicated that the history of antecedent therapeutic irradiation to the neck area was significantly very common (up to 80 per cent) in children and adolescents with thyroid cancer.*® ** In the ABCC survey in Japan, 21 instances of thyroid cancer were found over a three-year period (1958-1961) among more than 19,000 persons in the study sample.*? Nineteen of the 21 occurred in the exposed group. Two were under 15 years of age at time of exposure; 8, however, were under 21 years of age at exposure. Tentative exposure doses (T;;), computed on the basis of exposure distance and shielding history of these cancer cases, ranged from 125 to 3,400 rads with the exception of a single patient whose T;; dose was calculated to have been 33 rads.®* It was concluded from these findings that thyroid carcinoma was significantly more prevalent among survivors \ | : : i ‘ af n! i Rest Pe Prat: oSi Qe“2Zu Cancer. The correlation between ionizing October 1965 |