Development of an artificial intelligence-based precision medicine technology that predicts immunotherapy response in cancer patients.
Immunotherapy is a new cancer treatment that uses the immune system to fight cancer cells instead of chemotherapy or radiotherapy. Because it uses the body's immune system to attack only cancer cells, it has fewer side effects than conventional anticancer drugs. Furthermore, because it uses the immune system's memory and adaptability, patients who have benefited from its therapeutic effects experience long-term anticancer effects.
The recently developed immune checkpoint inhibitor has significantly improved the survival rate of cancer patients. However, the problem with cancer immunotherapy is that only about 30% of cancer patients benefit from it, and current diagnostic techniques do not accurately predict the patient's response to the treatment.
In this context, the research team at POSTECH led by Professor Sanguk Kim (Department of Life Sciences) is gaining attention for improving the accuracy of predicting patient response to immune checkpoint inhibitors (ICIs) by using network-based machines
learning. The researchers discovered new network-based biomarkers by analyzing clinical data from over 700 patients with three different cancers (melanoma, gastric cancer, and bladder cancer) as well as transcriptome data from the patients' cancer tissues.
The team successfully developed artificial intelligence that could predict the response to anticancer treatment by utilizing network-based biomarkers. The researchers also demonstrated that treatment response prediction using the newly discovered biomarkers was superior to that using traditional anticancer treatment biomarkers such as immunotherapy targets and tumor microenvironment markers.
The researchers previously developed machine learning to predict drug responses to chemotherapy in patients with gastric or bladder cancer. This study demonstrated that artificial intelligence using gene interactions in a biological network could successfully predict patient response to chemotherapy as well as immunotherapy in multiple cancer types.
This study aids in the early detection of patients who will respond to immunotherapy and the development of treatment plans, resulting in personalized precision medicine and more patients benefiting from cancer treatments. This study was recently published in Nature Communications, an international peer-reviewed journal, and was supported by the POSTECH Medical Device Innovation Center, the Graduate School of Artificial Intelligence, and ImmunoBiome Inc.
Reference: DOI: 10.1038/s41467-022-31535-6