Cancer
Lung cancer risk update in the US Radiologic Technologists Cohort: Can we revise sex-specific radiation-related risk estimates at low dose-rates? Cato M Milder* Cato Milder Elizabeth K Cahoon Martha S Linet Dale L Preston Bruce H Alexander Cari M Kitahara
NASA risk projections show female astronauts are more likely than males to die from radiation-induced cancer. This finding stems from studies of Japanese atomic bomb survivors, which estimated greater radiation-related risk of lung cancer in females than males, though other cohorts have not replicated this sex-specific difference. The US Radiologic Technologists (USRT) cohort provides an opportunity to assess sex-specific lung cancer risk in a large occupational cohort with detailed smoking information.
Included were 72,238 female and 22,293 male US radiologic technologists who responded to a baseline survey and were alive and cancer-free at follow-up start. Lung cancer incidence from 1999-2013 was obtained from linkages with 43 state/regional cancer registries. Cumulative lung doses (mean=13.4 mSv for females, 17.9 mSv for males) were estimated through 1997. We used Poisson regression to estimate excess relative risk (ERR) of lung cancer per 100 mGy, stratified on age, birth year, and sex, adjusting for time-varying smoking status (ever/never/former) and pack-years. We tested dose-response effect modification by sex.
A total of 865 and 372 lung cancers occurred in females and males, respectively, during follow-up (median 14 years). We observed increased lung cancer risk with greater cumulative lung dose [ERR/100 mGy: 2.00, 95% CI: 0.51, 5.31]. In sex-specific models, the male ERR (3.92, 95% CI: 0.71, 10.28) was higher than the female ERR (0.11, 95% CI: -0.33, 2.26) (p-interaction: 0.45).
We did not observe a stronger positive association between cumulative lung dose and lung cancer incidence in females than males. However, this analysis contains considerable uncertainty, particularly as doses are only estimated through 1997, and these results could not be replicated in USRT lung cancer mortality analyses. Nonetheless, our results challenge NASA’s assumptions of elevated lung cancer risk in females, suggesting the need to combine multiple cohorts for risk projection.