Artificial intelligence is beginning to offer many benefits, particularly in the healthcare industry. In the field of cancer research, for example, scientists recently developed a program that can predict how cancer will evolve.

The new program is called REVOLVER (Repeated evolution of cancer). It was created by a team led by scientists from The Institute of Cancer Research in London and the University of Edinburgh. To create it, researchers analyzed the genomic structure of 768 tissue samples from the tumors of 178 patients with bowel, lung, breast, or kidney cancer. By studying the genetic patterns of the samples, researchers were able to identify characteristics that predict how a tumor will progress over time.

One of the most difficult problems of cancer is, of course, when it spreads beyond the original site to other organs or places in the body. This process is almost always unpredictable, and it is why prognosis is generally better for patients with early stages of the disease, which has not yet spread and can therefore be surgically removed, or treated with methods such as radiation therapy.

It is also often difficult to predict how a patient’s cancer will respond to treatments. Even if a patient has the same type of cancer as another, such as pancreatic cancer, differences in the tumor’s unique genetic pattern mean that it could respond completely differently to the same treatment. Some tumors even grow resistant to drug treatments, and when this occurs, options become limited. REVOLVER makes it possible for doctors to analyze the genetic pattern of a patient’s tumor, and select treatments that are the most effective for that particular tumor type, allowing them to proactively attack cancer in ways they know are more likely to work.

AI makes it possible for doctors to analyze the genetic pattern of a patient’s tumor, and select treatments that are the most effective for that particular tumor type, allowing them to proactively attack cancer in ways they know are more likely to work.

For example, if a tumor has a genetic pattern that predicts it might become resistant to treatment, doctors could use more aggressive strategies earlier on to try to eradicate the disease more rapidly.

The program can also make predictions about how long patients with certain types of tumors may survive. For example, patients with breast tumors that had a specific error in a protein called tumor-suppressing protein p53 (TP-53) (one of the most common mutations found in human cancers) as well as mutations in chromosome 8, were found to have a shorter survival time than other patients with similar tumors.

While this is not the first study to examine the genetics of tumors, it is the first time such research has been done on a large scale, which is important. When a tumor grows and changes, tumor cells multiply and reproduce just as normal cells would. During this process, the tumor develops mutations, or changes in its genetic pattern. The problem is that when analyzing the genetics of a single tumor or a small group of samples, it’s not always possible to determine which changes are important and are signals of how the tumor will grow, and which do not actually have an effect on the tumor progression. With such a large sample to study, scientists were able to pinpoint characteristics in multiple samples, that predicted with a high level of accuracy how the cancer would progress.

As the use of AI increases, concerns have been raised about issues such as data privacy and automation taking over human jobs, particularly in pathology, or analyzing samples to make a diagnosis–the area that REVOLVER itself has made progress in–as well as radiology, or medical imaging.

for now, REVOLVER represents an important step forward in cancer care. As joint study leader Professor Guido Sanguinetti, from the School of Informatics at the University of Edinburgh, said

“This study shows how collaboration across disciplines adds value to research. By solving a statistical machine learning problem, we were able to shed light on cancer evolution. It is an example of how the power of AI to detect complex patterns in data can be harnessed to further our scientific understanding to improve human health.”

Guido Sanguinetti