Interlaboratory Comparison of Dicentric Chromosomes Assay using electronically transmitted Images

ROY L.; GARCIA LIMA O.; STUCK OLIVEIRA M.; MENDEZ ACUÑA L.; DI GIORGIO M.; VALLERGA M.B.; RADL A.; TAJA M.R.; BUBNIAK R.; SEOANE A.; DE LUCA J.C.; MARTINEZ LOPEZ W.; DI TOMASSO M.; LAMADRID A.; GONZALEZ MESA J.; ROMERO AGUILERA I.; MANDINA CARDOSO T.; GUERRERO CARVAJAL C.; ARCEO MALDONADO C.; VALDIVIA P.; ESPINOZA M.; OLIVEROS N.V.; BARQUINERO F.

Mandelieu La Napoule

Congreso; ERP Biodose 2010; 2010

European Radiation Research Society, European Radiation Dosimetry Group

The increased risk of terrorism attacks has stimulated the scientific community to develop tools that would help the medical doctors to treat victims. In the case of radiation exposure the estimation of dose is valuable information for the medical team. This can be achieved by the dicentric assay in lymphocytes. However this approach is time consuming and the capacity of each individual laboratory performing the assay is limited. The bottleneck in data acquisition is the need to score dicentrics in a large number of cells. One way to increase the capacity of a given laboratory is to use the ability of skilled operators of other laboratories. This can be done using image analysis systems and distributing images all around the world. Two intercomparisons involving the Latin American biological dosimetry network and two European laboratories were conducted during 2009 to test the efficiency of such an approach, in the frame of the IAEA Regional Project-RLA 9/061. For each trial one laboratory was responsible for all the process from blood irradiation to supplying images through a website created by the network. After performing image analysis the various laboratories sent their results to one nominated person who was responsible for the analysis and evaluation of the results. During the first intercomparison, 9 laboratories analyzed the same 100 images derived from cells exposed to 0.5 Gy and another 100 images derived from cells exposed to 3 Gy. Whatever the dose only around half of the cells were suitable for dose estimation. From this first intercomparison a coefficient of variation (CV) of 20 % was obtained for both doses using two end-points: the measured dicentric frequency and the estimated dose. The trueness, which is the closeness of agreement between the average value obtained from the participant laboratory results and an accepted reference value, was evaluated. This trueness was very good for the 3 Gy dose (3 %) but much less (60 %) for the 0.5 Gy dose. However, at this level of dose this poor agreement does not have big impact on health consequences. In the second intercomparison (0.5 Gy exposure) an emergency situation was tested and each lab was required to score 50 different images in 2 days extracted from 500 downloaded images. Then they had to score the remaining 450 images in a week. Among the 8 participants these requirements were achieved by only 3 laboratories. These three were those where there was more than one skilled operator. Using 50 different images, the CV of frequency (75%) and dose (70%) were not as good as in the first intercomparison where the number of images was 100. At this stage it is difficult to say if these higher CV derive from the lower number of cells analyzed (50 vs. 100) or from the fact that not the same set of cells were scored. The trueness was better after scoring 500 cells than after 50 cells. The results obtained so far with images analysis seem promising and have to be validated as a technique for population triage in a large scale accident.