A group of researchers from the University of Helsinki, Finland, has developed a tool based on open-source artificial intelligence (AI) to detect the immune response of an individual to the treatment of skin cancer.
Using the AI model, the researchers counted the number of immune cells that responded to cancer therapy. Their quantity was seen as a direct pointer to how people with skin cancer would respond to immune therapy targeting the cancer.
“In practice, the AI model makes it possible to diagnose skin cancer with a blood test, determine the prognosis and target therapies increasingly accurately,” the university said in a 2022 release.
The study also involved HUS Comprehensive Cancer Center, Aalto University and Stanford University. The researchers used the AI tool developed by Mark M Davis, a professor and director of the Stanford Institute for Immunology, Transplantation, and Infection.
The study comprised more than a thousand samples from 15 data sets totalling more than ten million T cell receptors from patients with melanoma and healthy individuals.
“Prior research has been unable to provide doctors with tools that would predict who will benefit from treatment that activates the defence system. The correct targeting of therapies is extremely important, since drug therapies are expensive and serious adverse effects fairly common,” said Dr Jani Huuhtanen, a doctoral researcher from the University of Helsinki and Aalto University, in a statement.
When the body does not respond to therapy
Typically, analysing how one responds to any medical treatment is essential to understand the underlying disease mechanism. If medical treatments are frequently ineffective, it indicates that the body is not responding to the therapy and that the immune system is not fighting the cancer.
Of the different types of T cells that the immune system has, only a few cells called anti-tumour T cells recognise the cancer antigens and respond to cancer therapies.
On the surface of T cells are TCRs or T cell receptors, which are proteins that act as binding hooks that latch on to antigens or cancer cells during an immune response.
The hypothesis of the study was that whichever individual’s body did not recognise the cancer cells as their enemy could not respond to the therapies.
Quantity is the key
The researchers sequenced T cell receptors (TCRs) and compared them with 783 healthy controls.
They found that individuals who had skin cancer had more immune cells than the healthy individuals.
Further individuals in whom the immune cells recognised the cancer cells responded to cancer therapies better than those lacking such differentiation.
“In future studies, our aim is to explore the utilisation of the AI model now developed and investigate whether it can predict treatment responses also for novel cancer drug therapies still in development,” says associate professor of computational biology and machine learning Harri Lähdesmäki from Aalto University.
“These AI-based technologies would be used immediately in western nations because they have more skin cancer cases diagnosed than the population of India.”- Dr Sunil Kumar Prabhu, consultant dermatologist, Aster RV Hospitals, Bengaluru, tells Happiest Health.
The AI model developed by these researchers also detected cancer-defensive immune cells in breast cancer, lung cancer and blood cancer.
The study could help us understand the antigen specific immune responses in individuals. It also allows us to understand the way our immune cells can be reprogrammed to fight against cancer antigens.