Performance Of The Dicentric Assay In A Recent NATO Exercise Of Established And Emerging Biodosimetry Methods

Accidents involving human exposure to radiation can cause severe health effects which may require extensive medical resources. Particularly in mass-casualty events, the rapid identification and classification of potentially overexposed individuals into medical treatment groups is of prime importance. For this purpose, clinical signs and symptoms and biological dosimetry methods are the two main approaches for assessing radiation exposure in situations where no dosimetry badge was worn. The dicentric chromosome assay (DCA) is considered the “gold standard” method for biological dosimetry after an acute radiation exposure. However, several novel techniques are emerging which may be faster and have a higher throughput than the DCA and could thus become valuable dosimetric tools in the future. This comprehensive study was organized under the umbrella of the NATO Research Task Group RTG-033 “Radiation Bioeffects and Countermeasures” in order to compare the performance of the two most validated techniques (DCA, cytokinesis block micronucleus assay) and of two candidate assays (-H2AX, gene expression) for biodosimetry. To this end, an inter-laboratory and inter-assay comparison exercise was performed. In a first step, blood samples exposed to known X-ray doses were provided for establishing calibration curves at each laboratory and for each assay. In a second step, ten coded blood samples irradiated with different X-ray doses were distributed among 15 institutions for triage-mode biodosimetry. This manuscript focuses particularly on the inter-laboratory comparisons of the DCA. Earliest dose estimates were reported only 2.4 d after sample receipt in the respective laboratory. An almost 7-fold difference in dose estimate precision (variance 0.07-0.47) was observed among participating laboratories. In particular, the calibration curve used and the actual dose levels of coded blood samples proved to be of significance in explaining the variances. Additionally, our analysis provided further hints to unused optimization potential of the DCA.