A Sensitive Blood Test for Rapid Estimation of Absorbed Radiation Dose

Abstract

Discovery of a panel of circulating miRNA biomarkers of which changes are dependent on radiation doses in rodent and non-human primate models as well as their validation in human AML/ALL patients receiving total body irradiation (TBI) allowed us to develop a novel radiation biodosimetry test. The available biodosimetry methods such as complete blood count (CBC) assessing lymphocyte depletion and dicentric chromosome assay (DCA) have limitations with regard to its feasibility, robustness and throughput. This novel assay allows rapid estimation of dose of exposure so that victims of mass casualty accidents could be triaged quickly. Male and female mice from different genetic background and age groups were subjected to TBI (0.5 to 10 Gy gamma rays, 137Cs source, 0.94 Gy/min). A single drop of blood was collected into a lysis solution; RNA was isolated and changes in miR150-5p and miR23a-3p were compared by qPCR using ΔΔCT method. The decrease in miR150-5p (blood cell origin) with miR23a-3p (not expressed at significant level in lymphocytes, but abundant in plasma) as internal reference were extrapolated to develop algorithms for dose estimation. Blinded studies and mathematical modelling were used for dose calibration and to evaluate the accuracy and the influence of variables such as age, gender and other potential confounders. Dose-dependent decrease in miR150-5p with miR23a-3p as internal reference was extrapolated that allowed dose estimation with ±0.5 Gy precision at 1 to 3 Gy and with ± 1 Gy precision up to 6 Gy at 24-96 hours, the dose range and time points relevant for triage (two-way ANOVA, p<0.05). Significant correlation was noted between miR150-5p/miR23a-3p assay and lymphocyte depletion kinetics (Pearson correlation coefficient r2 0.83; p<0.0001), confirming the reliability of new molecular biodosimetry assay. Algorithms were developed for dose reconstruction and the blinded studies showed precision in dose estimation (1% uncertainty). Linear regression model was developed for each time point indicated a good agreement between actual and calculated doses (r2>0.9, slope>=0.9, lack of fit test p>0.05). Robustness and sensitivity of the assay has been validated by comparing the values in human healthy volunteers and in leukemia patients prior, during and after radiation treatment and on marrow reconstitution after hematopoietic stem cell transplantation (HSCT). We have developed an internally controlled and simple two miRNA biomarker blood test for radiation biodosimetry with broad translational utility. The assay demonstrates superior capability in dose estimation and distinguishing the unexposed from 2 Gy exposed individuals (triage criteria) using a drop of blood that can be collected by finger-stick. Furthermore, our data demonstrate the utility of the assay in clinical radiation oncology, especially for evaluation of the marrow ablation following radiation and its reconstitution after bone marrow transplantation.