Bone Sarcomas Following Radiation Therapy

That ionizing radiation could induce neoplasia became evident soon after the discovery of roentgen rays. As early as 1907, Porter and White (18) collected 11 cases of skin carcinoma in radiation workers following chronic exposure. In 1910, Marie, Clunet, and Raulot-Lapointe (14) reported the occurrence of a soft-tissue sarcoma in a rat with radiation ulcer. Simultaneously, in 1922, Beck (1) and Marsch (15) reported bone sarcoma in irradiated patients. By 1956, Sabanas, Dahlin, Childs, and Ivins (20) were able to collect 70 cases of bone sarcoma induced by irradiation. Although the earlier human tumors were in tuberculous bones, Hatcher's (9) cases showed that radiation could also cause sarcomas in healthy bone. Since induction of bone sarcoma is a complication of radiation therapy, it seemed of interest to determine the incidence of such tumors. Previous reports have generally included the number of sarcomas seen at a particular institution, but not the percentage of treated patients in whom the tumor developed. In this study, the number of patients in whom sarcoma occurred is compared to the number treated. Material All records coded as bone tumors since 1928 in the Radiation Therapy Department, University of California Medical Center (San Francisco), were reviewed. The names of patients with bone tumors filed in the hospital record room were then compared with the therapy master file to identify patients who were both irradiated and suffering from bone tumors. This search revealed 2 who were irradiated at the University of California Medical Center and who met the criteria for radiation-induced bone sarcoma outlined by Cahan and his co-workers (3). To serve as a basis of estimating incidence, the files on all malignant diseases treated between 1928 and 1958 in the Radiation Therapy Department were reviewed. Skin carcinomas were excluded because of the limited area treated and the superficial type of radiation used. Except for the bone tumors, all benign diseases were excluded because of the low-radiation doses used in such patients. The number of radiation-treated patients and the number of five-year survivals were determined for each type of tumor. These five-year survival figures are thought to give a reasonable indication of the number of patients at risk. Actually, most of the five-year survivors had controlled disease and, therefore, have lived and been observed much longer. A total of 5,900 patients suitable for inclusion in the study were treated during this period; 2,300 of these were alive five years later. Included in the 5,900 patients are 1,011 with cancer of the breast and 1,204 with cancer of the cervix. Case Reports Case I. A 35-year-old woman, when first seen at the University of California Hospitals in 1943, gave a history of vaginal bleeding of four months duration.