2024-07-15 17:12:57
(ANSA) – MILAN, JULY 15 – Thanks to Artificial Intelligence, new markers have been identified, real “alarm bells”, to predict the risk of prostate cancer recurrence, even ten years later: this is the result of a new study conducted by researchers from the Human Technopole, the Institute of Cancer Research in London and the Royal Marsden NHS Foundation Trust.
The researchers – reads a press release from HT – have identified, thanks to the use of big data and Artificial Intelligence, that the co-presence of tumor cells with different genetic characteristics within the same tumor and differences in their shape, size and structure is indicative of the ability of the neoplasm to change over time. This evolutionary capacity is associated, even after a very long period of time, with a high risk of the disease returning.
The study, it is explained, could help doctors better personalize prostate cancer treatment, adopting more aggressive treatments in those cases in which, thanks to these parameters, a greater risk of recurrence emerges. The research, published today in the scientific journal Nature Cancer, differs from other works for the high number of samples analyzed and for having examined the disease in different stages of its development. Using machine learning, the researchers analyzed 1,923 samples from 250 patients, focusing on the spatial structure of the tissue. They also used a specially created AI technique to perform the Gleason classification, a scoring system that classifies cancerous tissue from one to five based on the pattern of its cells.
“In addition to producing better prognostic biomarkers for prostate cancer, our study provides further evidence of the predictive possibilities that come from studying how a single tumor evolves and changes over time,” explains Andrea Sottoriva, head of the Computational Biology Research Center at Human Technopole and corresponding author of the study.
Marino Zerial, director of the Human Technopole, underlines that “it is still an experiment and not a clinical practice but in the future this approach could help doctors to systematically classify patients based on the risk of disease recurrence and decide which therapies to adopt”. (ANSA).
2024-07-15 17:12:57