Uranium carcinogenicity in humans might depend on the physical and chemical nature of uranium and its isotopic composition: results from pilot epidemiological study of French nuclear workers.

OBJECTIVE

To study the cancer risk related to protracted, low-dose exposure to different industrial uranium compounds, paying attention to their isotopic composition and solubility.

METHODS

Two thousand and ninety-seven workers employed at the AREVA NC uranium processing plant (France) were followed up for mortality from 1960 to 2006. Historical exposure to uranium and other carcinogenic chemical and physical pollutants was assessed on the basis of the plant-specific job-exposure matrix. For each type of uranium, Cox regression models stratified on sex and calendar period, and adjusted for socioeconomic status and potentially confounding co-exposures were used to estimate hazard ratios (HRs) for mortality from lung cancer (53 deaths) and lymphatic and hematopoietic tissue malignancies (21 deaths).

RESULTS

We observed that exposure to reprocessed uranium entails increasing risks of mortality from lung cancer and lymphatic and hematopoietic malignancies (the most significant HR being respectively 1.14 (95% CI: 1.00-1.31) and 1.20 (95% CI: 1.01-1.43) per unit of a time-lagged log-transformed continuous exposure scores), and that the HRs tend to increase with decreasing solubility of the compounds.

CONCLUSION

Our results suggest that uranium carcinogenicity may depend on isotopic composition and solubility of uranium compounds. This study is the first to show the carcinogenic effect of slowly soluble reprocessed uranium on two uranium target organs. This finding is consistent with data from epidemiological and experimental studies on similar compounds but need to be confirmed in the more powerful dose-response analysis.