Indications of the neutron effect contribution in the solid cancer data of the A-bomb survivors.

Risk estimates for radiation-induced cancer are primarily based on the follow-up of the Japanese A-bomb survivors. Their exposures were due to gamma rays and neutrons, and, currently--with the assumed low RBE = 10 of neutrons and reference to the colon dose--the late radiation effects are almost fully attributed to the gamma rays. Solid cancer risk estimates for different organ sites are assessed here, and an inconspicuous but statistically highly significant trend of larger values is found for the organs closer to the body surface; i.e., the organs with less body shielding and, therefore, with larger neutron dose-fractions. Underestimation of the RBE of neutrons can explain this apparent correlation. The trend of ERR/Gy vs. depth ceases to be statistically significant for RBE values close to 100. The suggestion of high RBE values and the corresponding reduction of gamma-ray risk estimates is found to be in line with log-likelihood computations in terms of AMFIT, which provide for the solid cancer mortality of the A-bomb survivors the minimum deviance for RBE = 100 with a 95% confidence lower limit of 25. The present assessment had to use the data made publicly available by RERF. In this form they contain city-, sex-, age-, and dose-categories, but--instead of a separate neutron-dose category--only the mean neutron dose for each data cell. The tentative conclusions that are here obtained should, therefore, be examined by a more definitive analysis, either in terms of grouped data with a separate classification of neutron doses or, ideally, in terms of person by person calculations to be performed at RERF with individually estimated neutron doses.