Scientists have used artificial intelligence to predict how cancers will evolve and evolve, allowing physicians to design the most effective treatment for each patient.
A team led by the Institute of Cancer Research London (ICR) and the University of Edinburgh developed a new technique known as Revolver (repeated evolution of cancer), which seeks patterns in DNA mutation within cancers and uses the information to help future predict genetic changes.
They said that the ever-changing nature of tumors is one of the biggest challenges in the treatment of cancer – with cancers often evolving into a resistant form.
But if doctors can predict how a tumor will evolve, they can take action to keep cancer on its way before it has had a chance to evolve or develop resistance, thereby increasing the patient's chances of survival.
The team also found a relationship between certain sequences of repeated tumor mutations and survival results.
This suggests that repeating patterns of DNA mutations can be used as an indicator of the prognosis, which help shape the future treatment.
For example, researchers found that breast tumors with a series of errors in the genetic material coding for the tumor suppressing protein p53, followed by mutations on chromosome 8, survived less time than those with other similar pathways of genetic changes.
The team developed a new machine learning technique that transfers knowledge about tumors in comparable patients.
This method identifies patterns in the order in which genetic mutations occur in tumors that are repeated both within and between patients' tumors, using the mutation pattern of one tumor to predict the tumor of another.
Researchers used 768 tumor samples from 178 patients reported in previous studies for lung, breast, kidney and colon cancer, and analyzed the data within each type of cancer to accurately detect and compare changes in each tumor.
By identifying repeat patterns and combining this with the current knowledge of cancer biology and evolution, scientists were able to predict the future trajectory of tumor development.
If it is determined that tumors with certain patterns develop resistance to a certain treatment, this new method can be used to predict whether patients will develop resistance in the future.
The research is published in the journal Nature Methods.
Dr. Andrea Sottoriva, who led the study and is team leader in evolutionary genomics and modeling at the ICR, said: "We have developed a powerful artificial intelligence tool that can predict the future steps in the evolution of tumors based on certain patterns of mutation that has hitherto been hidden in complex datasets.
"With this tool we hope to remove one of the trump cards of cancer – the fact that it evolves unpredictably without knowing what will happen next.
"By giving us a look into the future, we could potentially use this AI tool to intervene earlier and predict the next step of cancer."
ICR CEO Paul Workman said: "The evolution of cancer is the biggest challenge we face when creating treatments that work more effectively for patients.
"If we are able to predict how a tumor will evolve, the treatment can be changed before adjustment and drug resistance ever occur, keeping the cancer one step ahead.
"This new approach with AI can make it possible to personalize treatment in a more detailed way and at an earlier stage than is currently possible, to adapt to the characteristics of each individual tumor and to predict how the tumor will develop in the future. will see. "