Chinese Researchers Develop AI Framework to Boost Immunotherapy

Select Language:

A team of Chinese researchers has developed a new computational system designed to identify potential combination immunotherapy compounds more efficiently, offering renewed hope for patients who have limited responses to current treatments.

This framework, known as IGeS-BS, was tested on over 10,000 compounds spanning 13 different types of cancer, including liver and colon cancers. It created an immuno-response landscape that helps prioritize candidates with the potential for synergistic effects, as detailed in a recent publication in a leading U.S. medical journal.

The system integrates transcriptomic data from thousands of patients who have undergone immunotherapy, successfully identifying 33 reliable signatures that predict immune response. These signatures are then used to develop a boosting score, which measures how each compound alters the tumor microenvironment. Based on these scores, compounds are ranked to pinpoint those most likely to improve treatment outcomes.

The research was led by experts from the Shanghai Institute of Nutrition and Health under the Chinese Academy of Sciences and Zhongshan Hospital affiliated with Fudan University.

Immunotherapy, especially the use of immune checkpoint inhibitors, has proven to be a highly effective cancer treatment that outperforms traditional methods. Usually, it is combined with other cancer drugs to enhance success rates. While current approaches rely heavily on expert knowledge and prior experience to find suitable drug combinations, the scientists suggest that computational predictions could streamline this process, shorten development times, and improve the chance of discovering effective drug pairings.

The market for cell-based immunotherapy in China is projected to grow by 41% this year, reaching approximately 22 billion yuan (around 3.1 billion USD). By 2030, it is expected to expand to 125 billion yuan (approximately 17.4 billion USD), driven by technological advances, according to ChinaIRN, a research organization.