The impact on radiation risk estimates of effect modifiers and confounders

We have conducted studies of the joint effects of radiation with other risk factors for liver, lung, and breast cancer in the cohort of Japanese survivors of the 1945 atomic bombings. Based on follow-up of 45,113 subjects from 1958 through 1994, of whom 592 developed lung cancer, we found the effects of smoking and radiation to be significantly non-multiplicative and consistent with additivity. Adjustment for smoking reduced the female:male ratio of radiation risk estimates for lung cancer in this cohort from 5.8 to 1.6, a ratio more similar to that for all solid cancers. We conducted cross sectional and case control studies within the A-bomb survivor cohort to assess the joint effects of radiation with hepatitis B virus (HBV) and C virus (HCV) infections on the etiology of liver cirrhosis and hepatocellular carcinoma (HCC). Our study of 268 pathologist-confirmed cirrhosis cases and 843 subjects without cirrhosis, found no relationship between A-bomb radiation and cirrhosis after adjustment for viral hepatitis. In terms of HCC, our study of 238 pathologist-confirmed cases and 894 controls showed super-multiplicative interaction between radiation and HCV infections. Our results suggest that while chronic radiation exposure acts as a complete carcinogen for HCC, acute irradiation may act in concert with an agent such as HCV that is associated with liver cell proliferation. Studies in progress of the joint effects of radiation with insulin-like growth factor 1 (IGF-1) and IGF Binding Protein 3 (IGFBP-3), and total estradiol on the etiology of breast cancer, a particularly radiogenic tumor, will also be discussed