Improved radiotherapy for cancer has produced a large population of long-term survivors. These survivors, however, appear to have a greater risk of second primary cancers than would be expected based on the incidence of first primary cancers in the general population. The etiology of radiation-induced cancer is multifactorial and incompletely understood, and there is no firm diagnostic criteria for radiation-induced cancer because there are no specific histopathological findings. The incidence of radiation-induced cancers following radiotherapy was estimated to be about 0.3% in 5-year survivors, based on mail surveys (in 1979 & 1984) in Japan. It seems that the benefit of treatment outweighs the risk of developing secondary tumors. However, we should avoid excess treatment of patients with Hodgkin's disease or many pediatric malignancies who will be able to survive a long time after treatment. The relative carcinogenicity of various combinations of radiation and chemotherapeutic agents is now being established. Chemoradiotherapy is now frequently used for treatment of various malignancies which are considered to be uncontrollable by chemotherapy or radiotherapy alone. We should minimize their carcinogenic effects and the risk of developing secondary treatment-related tumors, to produce an optimal therapeutic gain. Long-term close follow-up is necessary for these patients.