In brief

Under MRI scanning, young mice exposed to radiation showed impaired brain structures in middle age, including regions affecting movement and memory.

© Pexels

The dark side of brain radiotherapy

3 Nov 2022

Non-invasive imaging technology reveals how radiation exposure in the brain, particularly early in life, damages delicate brain tissue.

The Hippocratic oath of “do no harm” remains the cornerstone of modern medical ethics. Unfortunately, not all clinical interventions for cancer patients completely align with this age-old philosophy. Take radiotherapy for example, which uses strong beams of energy to destroy tumours but also leaves serious and lasting imprints on patients’ lives: many adults and almost all children who receive radiotherapy experience learning and memory problems after receiving radiotherapy.

While neuroscientists have documented these functional changes, the consequences of brain radiation exposure on a cellular level have remained unclear. For Bhanu Prakash, a Principal Investigator at A*STAR’s Bioinformatics Institute (BII), imaging technologies that allow scientists to look inside the brain non-invasively in real time may hold the key.

“Magnetic resonance imaging, or MRI, allows us to look into brain development systematically,” Prakash explained, speaking on a study he led using MRI to track the effects of radiation exposure in mice. “We can see how the brain structure and networks change in the same cohort of animals over a certain period.”

In collaboration with the Singapore Nuclear Research and Safety Initiative, Prakash and his team worked towards developing novel MRI-based biomarkers of brain damage to better monitor the long-term effects of early life radiation exposure.

Animals were divided into separate groups that received a single radiation dose at three, 10 or 21 days after birth. The researchers then performed MRI scans at 13 months or the ‘middle age’ in the life cycle of mice.

The team used a custom in-house bioimaging pipeline to precisely align the MRI scans to a reference atlas of the mouse brain. The innovative technology significantly sped up image analysis by automatically labelling regions of the brain and calculating any changes in brain volume.

“This approach allows us to not only identify the brain structures that are affected by radiation but also understand the strength and duration of radiation exposure at which the structural changes become irreversible,” said Prakash.

After 13 months, MRI scans found mice exposed to acute radiation at an earlier age had lower volumes of brain tissue in multiple regions compared to unexposed individuals.

© A*STAR Research

The data collected revealed widespread brain shrinkage and neuron death in multiple regions, including the cerebellum which maintains balance and movement and the hippocampus, the brain’s memory centre.

Furthermore, the researchers found that the earlier the exposure took place, the greater the extent of the structural damage in the brain—mice exposed to radiation three days after birth showed the highest impact, findings that lined up with previous clinical studies.

Prakash said that this proof-of-concept study highlights MRI-based approaches as a reliable, non-invasive alternative to current biopsy-based analytical methods. “We established that MRI can give very similar insights to the results obtained at the microscopic or histological level.”

The A*STAR-affiliated researchers contributing to this research are from the Bioinformatics Institute (BII) and the Institute of Bioengineering and Bioimaging (IBB).

Want to stay up to date with breakthroughs from A*STAR? Follow us on Twitter and LinkedIn!


Ren, B.X., Huen, I., Wu, Z.J., Wang, H., Duan, M.Y., et al. Early postnatal irradiation-induced age-dependent changes in adult mouse brain: MRI based characterization. BMC Neuroscience 22, 28 (2021). | article

About the Researcher

View articles

Bhanu Prakash KN

Principal Investigator

Bioinformatics Institute (BII)
Bhanu Prakash KN received his master’s and doctoral degree from the Indian Institute of Science in Bangalore where he investigated image processing algorithms for improving medical diagnosis. He joined Kent Ridge Digital Labs as a research scientist which later became part of A*STAR. He is currently a principal investigator at the Bioinformatics Institute where he leads the clinical data analytics and radiomics group. His research interests revolve around medical image processing, pattern recognition, biomedical signal processing, machine learning and medical instrumentation.

This article was made for A*STAR Research by Wildtype Media Group