Automated dose-effect calibration curve for X-rays using the cytokinesis-block micronucleus assay

This article shows the development of a dose–effect calibration curve for X-ray exposures ranging from 0 to 4 Gy using the cytokinesis-block micronucleus assay and automated analysis—the first effort of its kind reported in Latin America. This work establishes a regional benchmark for high-throughput methodologies in cytogenetic biodosimetry, highlighting their potential to improve operational efficiency and reduce response times in radiological emergencies. Methods: Blood samples from six healthy donors were irradiated with X-rays at seven dose levels (0–4 Gy) using a calibrated 6 MV linear accelerator. Two blind samples (1.5 and 3 Gy) were included for validation. The CBMN assay was performed following IAEA protocols, DAPI-stained slides were analyzed using a AxioImager.Z2 automated microscope integrated with MetaSystems Metafer4 and the MNScoreX classifier software. A negative binomial regression model (NB1) was used for model fitting, accounting for overdispersion in micronucleus (MN) frequency. Results: Automated scoring of binucleated lymphocytes showed a dose-dependent increase in MN frequency. The fitted model followed a linear–quadratic relationship: Y = 0.0545 + 0.0448·D + 0.0145·D², with all coefficients statistically significant (p < 0.001). Dose estimates for blinded samples (1.5 and 3 Gy) matched the true doses within 95% confidence intervals, with all z-scores < |3|. Conclusions: The resulting linear–quadratic dose–response curve enabled accurate estimation of blinded sample doses, with all z-scores falling within acceptable fitness-for-purpose thresholds. These results underscore the value of combining automated microscopy with robust statistical modeling to achieve reliable dose assessment, particularly in high-throughput settings and radiological emergency scenarios.