A comparison of the degree of curvature in the cancer incidence dose-response in Japanese atomic bomb survivors with that in chromosome aberrations measured in vitro

Abstract

Purposes : To compare the degree of curvature in the dose-response for chromosome aberrations and for radiation-induced cancer. Materials and methods : Comparison of the ratio of the quadratic and linear coeffcients (in dose) in Japanese atomic bomb survivor cancer incidence data, based on follow-up to 1987 and taking account of random errors in DS86 dose estimates with the same ratio in four datasets of chromosome aberrations in peripheral blood lymphocytes measured in vitro and exposed to 60 Co gamma radiation. Results : There are no statistically significant differences between the four in vitro datasets in the ratio of the quadratic to the linear coeffcients for dicentrics or chromosome translocations, nor are there indications of differences between this ratio for dicentrics and that for complete chromosome translocations (p >0.1 in all cases). If the 0-4 Gy dose range is used in the Japanese atomic bomb survivor data, the ratio of the quadratic to the linear coeffcients for all solid cancers is 0.06 Sv 1 (95% CI 0.22, 0.67) and so is not significantly different from 0; this ratio is statistically highly inconsistent (p <0.0001) with the analogous ratio estimated for the in vitro chromosome aberration data (4.20Sv 1; 95% CI 3.06, 6.51). By contrast, there are no statistically significant differences (p =0.42) between the ratio of the quadratic to the linear coeffcients for leukaemia incidence in the Japanese cohort, 1.81 Sv 1 (95% CI 0.21, > 1000), with that for chromosome aberrations in vitro. These results are not markedly changed if the 0-2 Gy dose range is used in the Japanese atomic bomb survivor data. Conclusions : It is unlikely that in vitro chromosome aberrations could be a correlate for the initiating radiogenic lesions leading to radiation-induced solid cancers. However, taken together with certain other biological information, it may not be unreasonable on this basis to use in vitro chromosome aberrations in peripheral blood lymphocytes as a correlate of the radiogenic lesions leading to radiation-induced leukaemia.