A team of international researchers has developed an artificial intelligence (AI) model that predicts which skin cancer patients will benefit from a treatment that activates the immune system. Medical Xpress reported that the AI model can be used to diagnose skin cancer with a blood test to determine the prognosis and identify the treatment best suited for them.
They published their study in the scientific journal Nature Communications.
A technician viewing a blot on a fluorescence microscope while another technician is using a pipette at the Advanced Technology Research Facility (ATRF), Frederick National Laboratory for Cancer Research, National Cancer Institute.
Recognizing Skin Cancer to Determine the Right Treatment
Boosting the body's immune system has proven to be an effective therapy against skin cancer, but the problem with existing therapies that activate the natural defense system of the body is that they differ between patient groups. Some patients are cured, but others did not benefit at all from the treatment.
Jani Huuhtanen from the University of Helsinki and Aalto University said that [previous studies provided doctors tools that would predict who will benefit from treatment that activates the immune system. That means it is important to accurately identify those patients who would respond better to therapies.
The team hypothesized that the immune cells of patients for whom the therapy is ineffective do not recognize that skin cancer is an enemy, Labiotech.eu reports.
They analyzed samples from almost 500 skin cancer patients and compared them to the samples taken from 1,000 healthy patients using their AI model. Then they used another AI model created by Mark M. Davis' lab at Stanford University to help them interpret the results. They were able to calculate the number of immune cells that recognized skin cancer.
As expected, the team found more skin cancer-sensing defensive cells in patients with melanoma than in healthy patients. Satu Mustjoki, a professor of translational hematology at the University of Helsinki, said that their findings could make it possible in the future to diagnose skin cancer from blood samples.
Researchers added that skin cancer patients with more defensive cells that recognize the cancer were more likely to benefit from the treatment that activates the immune system compared to those without those cells.
Using AI in Medicine
In a similar report from News Medical Life Sciences, researchers pointed out the significance of using AI models in the field of medicine. Over the past years, AI in medicine has grown exponentially and its application requires a long-term collaboration between doctors and researchers specializing in AI.
Harri Lähdesmäki, an associate professor of computational biology and machine learning at Aalto University, stated that they aim to explore the uses of the AI model in their future studies. Now, they were able to develop and investigate whether it can predict treatment responses on novel cancer drug therapies that are under development.
They have made their AI model available to other researchers and doctors for free by basing it on open-source software. In that way, they would also encourage its future development.
RELATED ARTICLE: T Cells Use Mechanical Force to Punch Through Tumors, Kill Cancer
Check out more news and information on Cancer in Science Times.
