Risk of cancer associated with low-dose radiation exposure: comparison of results between the INWORKS nuclear workers study and the A-bomb survivors study

Klervi Leuraud,Monika Moissonnier,Michael Gillies,Ausrele Kesminiene,Richard Haylock,Mary K Schubauer-Berigan,Elisabeth Cardis,Robert D Daniels,David B Richardson,Dominique Laurier,Isabelle Thierry-Chef

doi:10.1007/s00411-020-00890-7

Abstract

The Life Span Study (LSS) of Japanese atomic bomb survivors has served as the primary basis for estimates of radiation-related disease risks that inform radiation protection standards. The long-term follow-up of radiation-monitored nuclear workers provides estimates of radiation-cancer associations that complement findings from the LSS. Here, a comparison of radiation-cancer mortality risk estimates derived from the LSS and INWORKS, a large international nuclear worker study, is presented. Restrictions were made, so that the two study populations were similar with respect to ages and periods of exposure, leading to selection of 45,625 A-bomb survivors and 259,350 nuclear workers. For solid cancer, excess relative rates (ERR) per gray (Gy) were 0.28 (90% CI 0.18; 0.38) in the LSS, and 0.29 (90% CI 0.07; 0.53) in INWORKS. A joint analysis of the data allowed for a formal assessment of heterogeneity of the ERR per Gy across the two studies (P = 0.909), with minimal evidence of curvature or of a modifying effect of attained age, age at exposure, or sex in either study. There was evidence in both cohorts of modification of the excess absolute risk (EAR) of solid cancer by attained age, with a trend of increasing EAR per Gy with attained age. For leukemia, under a simple linear model, the ERR per Gy was 2.75 (90% CI 1.73; 4.21) in the LSS and 3.15 (90% CI 1.12; 5.72) in INWORKS, with evidence of curvature in the association across the range of dose observed in the LSS but not in INWORKS; the EAR per Gy was 3.54 (90% CI 2.30; 5.05) in the LSS and 2.03 (90% CI 0.36; 4.07) in INWORKS. These findings from different study populations may help understanding of radiation risks, with INWORKS contributing information derived from cohorts of workers with protracted low dose-rate exposures.

Highlights

Radiation protection guidelines are informed by evidence derived from cellular studies and animal experiments (UNSCEAR 2008)
The periods of follow-up and ages at exposure are similar between the two populations; the study populations include people first exposed at ages 20–59 years and the mean age at time of bombings in the Life Span Study (LSS) is 37.3 years, while the mean age at mid-career exposure is 37.7 years
In the current paper a set of parallel analyses of cohort mortality data was conducted for members of the Japanese atomic bomb survivors included in the LSS (Ozasa et al 2012) and nuclear workers included in International Nuclear Workers Study (INWORKS) (Laurier et al 2017) considering mortality due to solid cancers and leukemia, outcomes of relevance to current risk models used in radiation protection

Summary

Introduction

Radiation protection guidelines are informed by evidence derived from cellular studies and animal experiments (UNSCEAR 2008). The most influential of these human studies is the Life Span Study (LSS) of Japanese atomic bomb survivors (Ozasa et al 2012). This epidemiological study has a number of features that have been noted as grounds for its widespread use in radiation protection across the world: it is a relatively large cohort study that includes males and females who were exposed to radiation across a range of ages and doses of varying magnitudes, and this cohort has been followed for over 60 years since the atomic bombings. The appropriate use of results from acute exposure settings for elaborating radiation protection guidelines in settings of protracted low dose-rate exposures remains an open issue (HLEG 2009; MELODI 2016; ICRP 2017)

Methods

Results

Discussion

Conclusion