Radiation risks of Russian liquidators of the Chernobyl accident for the period 1992-2017. Part II: solid cancer mortality

This paper presents results of the assessment of radiation risks of mortality from solid malignant neoplasms (MNs) in the cohort of Russian emergency workers involved in the recovery operations following the Chernobyl accident (liquidators) for the follow-up period 1992-2017. The research was carried out at the National Radiation and Epidemiological Registry (NRER). The cohort comprises 69,440 males entered the Chernobyl exclusion zone from 1986 over 1987, summed person-years of observation - 1,292,978, average age at arrival in the zone was 33.9 years, average whole-body external gamma-radiation dose, cumulated for the period of work in the exclusion zone was 132.9 mGy. For the follow-up period 4,272 deaths from solid cancers (ICD-10: C00-C80) were registered. Statistically significant estimates of excess relative risk (ERR) of mortality from all solid cancers per unit dose (ERR/Gy) was calculated with the linear non-threshold (LNT) model, the aggregate ERR/Gy=0.63; 95% confidence interval (CI) (0.19; 1.11); p=0.004). The aggregate radiation risk of mortality from 994 esophageal, gastric and colorectal cancers (ICD 10: C15-C16, C18-C21) and 1,484 tracheal, bronchial and lung cancers (ICD-10: C33-C34) were ERR/Gy=0.87 (95% CI: -0.01; 1.99) and ERR/Gy=0.60 (95% CI: -0.11; 1.47), respectively. Aggregate non-parametric estimate of relative radiation risks (RR) from all solid cancers, calculated in the dose intervals, were in agreement with the risk assessed with the LNT ERR model, taking into account 95% CI of these estimates. The age-related mortality rates in the cohort of liquidators did not differ from the corresponding rates in the male population of Russia. The standardized mortality ratio (SMR) for the follow-up period 1992-2017 was SMR=1.00 (95% CI: 0.97; 1.03). Further study of the cohort of liquidators will make it possible to assess radiation risks of mortality from cancers of different sites and to determine the statistical acceptability of the LNT risk model in the low-dose range (up to 100 mGy).