Scientists are using artificial intelligence to reveal previously invisible patterns inside breast cancer tumours, which could help doctors predict how the disease might progress.
CenSegNet platform developed
The technology could pave the way for more personalised, breakthrough treatments in tackling cancer. Centrosomes – tiny structures that help divide DNA evenly when a cell makes a copy of itself – have been a “hallmark of cancer” for more than a century, according to experts at the University of Southampton.
However, studying them in tumour tissues has proved “extremely challenging” because of their tiny size and constant changing. In cancer, centrosomes multiply too much, driving the progression of the disease.
Study findings
The AI platform, known as CenSegNet, was developed to analyse hundreds of thousands of cells in tumour samples. The study, published in Nature Communications, included tissue from 127 breast cancer patients being treated at University Hospital Southampton.
Researchers analysed more than 330,000 centrosomes, with CenSegNet uncovering two distinct abnormalities which had previously been considered as part of the same process. One involved cells acquiring too many centrosomes while the other had abnormally enlarged centrosomes.
Researchers found these defects behaved independently and can occupy different areas of a tumour.
Expert comment
Dr Salah Elias, of the University of Southampton’s school of biological sciences and institute for life sciences, said: “For more than a century, centrosome abnormalities have been recognised as a hallmark of cancer, but studying them in patient tissues has been extremely challenging.
“CenSegNet allows us to analyse these defects at single-cell resolution across entire tumours and uncover patterns that were previously impossible to see.
“Rather than viewing centrosome abnormalities as a single phenomenon, our study shows that they have distinct biological states with different spatial distributions and clinical associations.”
Clinical implications
The platform also helped uncover a link between different centrosomes and features of cancer. Tumours with high levels of enlarged centrosomes were more aggressive, and patients whose cells had lower levels had a better chance of survival.
Dr Elias said: “Specific combinations of defects may influence how a tumour grows, invades surrounding tissues and responds to treatment.
“This opens the door to developing new biomarkers and, ultimately, more personalised treatment strategies.”
Researchers are hopeful AI could one day be used to track disease by analysing the behaviour of cell structures. The team is also planning to combine CenSegNet with more data to explore if it can help guide treatment decisions.
