NEW YORK (GenomeWeb) – A serum-based microRNA signature might be able to distinguish people who have been exposed to radiation, such as through war, a terrorist attack, or an industrial accident, and predict whether they will die from that exposure.
Researchers from the Dana-Farber Cancer Institute and elsewhere identified a set of microRNAs that differed in mice and non-human primates that were or were not exposed to radiation, as well as another set of miRNAs that varied between mice and non-human primates that survive and did not survive that exposure. The researchers, who reported their results recently in Science Translational Medicine, noted that these miRNAs are evolutionarily conserved from mice to non-human primates to humans, suggesting that these signatures could also be useful in humans.
Currently, the dose of radiation a person has received is gauged by the time it takes them to develop radiation sickness, the kinetics of lymphocyte depletion, and an analysis of chromosomal abnormalities.
As supplies of radiation-mitigating drugs are limited, being able to triage people who received lethal versus sub-lethal doses and determining who will benefit from these drugs is a key challenge, said Dipanjan Chowdhury, a principal investigator at Dana-Farber and senior author of the study.
"I really wanted to establish that [miRNAs] can be useful as biomarkers [and] to develop a blood test that could give first responders some clue about how much a person got exposed to," Chowdhury said in an interview.
However, how applicable the signature will be to real-world scenarios is unclear.
In previous work, also published in Science Translational Medicine, Chowdhury and his colleagues found that serum miRNAs could distinguish mice given sub-lethal and lethal doses of radiation within 24 hours of exposure. Up until about three weeks after exposure, Chowdhury said, the mice were otherwise indistinguishably sick, but then the mice exposed to the sub-lethal dose — 6.5 grays — made it. He added that no test could tell the two cohorts apart, but that his group's miRNA signature could tell which mouse would live and which would die after radiation exposure.
"So, then, the question obviously becomes how useful this is in humans?" he said.
By reaching out to his colleagues at the Armed Forces Radiobiology Research Institute, Chowdhury obtained blood samples from non-human primates to examine such an miRNA signature in primates. He noted that these animals had been previously irradiated and their samples stored and that he and his team didn't conduct any new experiments on non-human primates for their study.
Blood was drawn from 48 rhesus macaques a week before they were exposed to total body irradiation and 24 hours after exposure. A portion of the macaques were treated before radiation exposure with the radioprotective agent gamma-tocotrienol (GT3). The macaques received 5.8 grays, 6.5 grays, or 7.2 grays of radiation, which have an expected lethality of 30 percent, 50 percent, and 70 percent in 60 days, respectively, without supportive care.
Some 25 miRNAs were either up- or downregulated in the monkeys after irradiation, the researchers reported. They noted that there was no significant dose dependence of the radiation-associated miRNAs, leading them to merge those sets into a composite set of samples. Three miRNAs — miR-133b, miR-215, and miR-375 — could perfectly distinguish the irradiated macaque samples from the non-irradiated ones, the researchers reported.
A classifier built on these three miRNAs had a sensitivity and specificity of 95 percent, they added.
These miRNAs, Chowdhury and his colleagues noted, are conserved from mice to rhesus macaques to humans. "Everything we got is completely conserved in humans. It's 100 percent conserved, the sequence is the same, so there's no reason to believe this is not going to work [in humans]," he said.
Similarly, Chowdhury and his colleagues identified two miRNAs — miR-30a and miR-126 — that were associated with survival in macaques. A classifier based on those plus gender, as female macaques appeared to be more sensitive to radiation, could predict radiation-induced mortality.
However, Naduparambil Jacob, an assistant professor of radiation oncology at Ohio State University who has also worked on uncovering miRNAs that act as biomarkers of radiation exposure, said he isn't sure how helpful such a classifier of survival would be in real-life situations. He noted that the classifier just gives a read-out of "die" or "not die," rather than a number, whereas in triage situations, clinicians want to know to what dose the patient was exposed to.
In addition, Jacob said the researchers didn't see a good dose-response and that they had to lump all the exposed macaques together to see this effect.
The source of the miRNAs and what their function may be is also unclear. Chowdhury noted that they could be coming from the bone marrow or the stroma and that they could be functional or the result of a secondary effect.
Some of the macaques Chowdhury and his colleagues studied had been treated with the radioprotective drug GT3. Serum levels of the miRNAs miR-30a, miR-126, and miR-375 in the treated macaques resembled that of macaques that did not undergo irradiation. This indicated to the researchers that these miRNAs could be used to gauge whether or not GT3 was having a protective effect.
Jacob noted, though, that these macaques received GT3 before they were irradiated, an order of events that would be unlikely to occur in the event of a nuclear attack or accident.
Still, Chowdhury and his colleagues combined the three radiation biomarker miRNAs and the two survival indicator miRNAs, along with two normalizer miRNAs, into a classifier that they said could detect radiation exposure and estimate the probability of death within 24 hours of that exposure.
"If this pans out, and I am not saying it will, but at least based on what we have so far… it will be cheaper, much more sensitive, and you would get real information in the week after exposure," Chowdhury said. "Those are huge advantages. But we need to really do more to make sure that it pans out."
In particular, he said, the classifier would need to be tested in a greater number of non-human primates and in samples, if they could be obtained, collected from people who were exposed to radiation at, for instance, Chernobyl or Fukushima.
He noted, though, that human samples might not be strictly necessary to develop the classifier as the radiation mitigator Neupogen gained FDA approval based on tests conducted in non-human primates.