Résumé : The first objective of our study was to analyse whether biomarkers for genotoxic effects (DNA breaks and alkalilabile sites and micronucleus and non-disjunction frequencies) could be fully validated for biomonitoring workers chronically exposed to ionizing radiation (IR). Blood samples of controls and individuals chronically exposed to IR were analysed. The interindividual variation was reduced when the comet data were adjusted for interexperimental variation, but remained statistically significant. No differences were found between groups, either for smoking or for exposure status. The second objective was to determine whether the Comet assay can be used to evaluate global repair phenotype as a susceptibility biomarker for IR-induced DNA damage in nuclear workers. A pilot study was performed and blood from workers exposed or not to radiation was submitted to a challenging dose of γ-rays. The repair kinetics of each individual donor were analysed by Comet assay at different time points and compared with the frequencies of biomarkers of genotoxic effects. There was a statistically significant interaction between biomarkers assessing the same damage (micronucleus and Comet assays). Multivariate analysis showed that micronucleus frequencies were positively influenced by age and the percentage of residual tail length was negatively influenced by the interaction between smoking and exposure status. The general conclusions from our study are: (i) a positive correlation exists between mechanistically related biomarkers; (ii) multivariate regression analysis confirmed that the interaction between smoking and exposure to IR negatively and statistically significantly influenced residual tail length; (iii) use of the Comet assay for the estimation of global repair phenotype with respect to IR is recommended because it is simple, fast and differences in in vitro repair capacity can be detected.