Between accidents, terrorism, and warfare, future nuclear exposures are virtually assured. From Hiroshima to Chernobyl, the probability is that some disasters are inevitable.

Survival of those exposed to external ionizing radiation will depend largely on identifying the extent of the dose, and treating accordingly.

A new meta-analysis looks at more than 10,000 genes and attempts to narrow down the window for accurate diagnosis to just two days, as presented in the journal PLOS ONE.

“Gene expression is a promising approach for radiation dosimetry assessment,” the authors said. “However, heterogeneity in protocols and analysis methods will require additional studies to confirm these results.”

The study looked at 24 papers published from 1978 through last year that analyzed radiation-induced expression of 10,170 genes, according to the paper.

The sample size was narrowed down to 31 genes that appeared in at least half of the studies assessed, the researchers said. Twenty-seven of them showed statistical significance through Spearman’s rank-order correlation methodology.

Five specific genes showed the best discriminatory factor: TNFSF4, FDXR, MYC, ZMAT3, and GADD45A. 

The combination of three of them at a time showed reliable interpretation, the researchers added.

“There is no validated gene signature to assess the radiation dose,” Jerome Lacombe, of the University of Arizona, said. “We hope this paper can begin to identify these biomarkers and confirm genes that are radiation responsive.”

The effective results could be crucial to future events that are unforeseeable, and particularly dangerous because of it.

“In the case of a nuclear event, a lot of people can be radiation. If you have thousands of people and only two days to screen everyone, it would be almost impossible to do this with the current exposure test,” Lacombe said. “With immediate care and the right diagnostics, people could have a better chance of survival.”

The study, which received funding from the National Institute of Allergy and Infectious Diseases, also featured authors from Texas A&M and Columbia University.