Purpose: Radiation-induced cancer has been epidemiologically investigated in occupational or atomic radiation exposure cases. These is also a need to clarify the risk in the case of radiotherapeutical exposure. In this study, therefore, cases of post-irradiated uterine carcinoma which was well cured and followed up over a long term were selected for statistical analysis of radiation-induced cancer in the medical division.
Materials and methods: A total of 19,384 patients with uterine carcinoma at seven institutions in Japan were registered on a computer system at the NIRS and were statistically analyzed. Of these, 6,655 patients were treated by surgical procedure alone, 4,310 were given a combined treatment modality of radiation and surgery, 8,419 were treated by radiation alone between 1960 and 1978. Radiation-induced cancer was defined according to the following categories: Secondary cancer was developed within the irradiated field. Time interval was over 5 years after the initial irradiation (leukemia was over 2 years). The cancer had a different histological type to the original one. A total of 43 patients with induced cancer were observed, namely: 14 with rectal cancer, 8 with leukemia, 6 with uterine corpus cancer, 4 with urinary bladder cancer, 3 with osteosarcoma or uterine sarcoma, 2 with sigmoid colon cancer or malignant fibrous histiocytoma, and 1 with ovarian cancer, respectively.
Results: Rectal cancer, leukemia and urinary bladder cancer were initially analyzed, because their expected values were easily estimated from the basic data. A total of 8,333 patients (43,418 person-years) from 7 institutions were considered. Their average follow-up period was 10.2 years after treatment. 1) Rectal cancer: Observed value (O) = 14, expected value (E) = 7.83, O/E = 1.79, 95% confidence interval of O/E ratio (CI) = 0.98-2.99, there was no significant difference. 2) Leukemia: O = 8, E = 3.78, O/E = 2.12, CI = 0.91-4.18, no significant difference. 3) Urinary bladder cancer: O = 4, E = 2.23, O/E = 1.79, CI = 0.49-4.55, no significant difference. In other words, no significant difference was observed in the risk of occurrence of secondary cancer among the total number of irradiated patients from the 7 institutions. Since these results, however, were influenced by the accuracy of patient follow-up, the O/E ratio of double cancer in each institution was observed to check the accuracy of the follow-up and 3 institutions were selected for further analysis comprising 2,686 patients (13,588 person-years) in all. Their average follow-up period was 10.1 years.(ABSTRACT TRUNCATED AT 400 WORDS)