Scientists in China have mapped out how genetic mutations allow tumors to build "immune barriers", a discovery that could explain why modern cancer treatments often fail in patients whose cancer has spread.

The study, published on Thursday in the journal Cell, offers a potential road map for making immunotherapy — a treatment that uses the body's own immune system to fight disease — more effective against metastatic tumors, which are the leading cause of cancer-related deaths.

The research was a collaborative effort between the Shanghai-based Center for Excellence in Molecular Cell Science, Shanghai Jiao Tong University, and the Guangzhou Laboratory. Using a new technological platform dubbed CLIM-TIME, the team analyzed 391 common genetic mutations to see how they reshape the tumor microenvironment — the biological neighborhood surrounding a tumor.

The researchers found that while some mutations make tumors easier for the immune system to attack, others create a physical fortress. Specifically, the loss of certain "tumor suppressor genes" — genes that normally prevent cancer growth — triggers a massive buildup of collagen.

This excess collagen makes the tumor structure so dense that it acts like a wall, preventing T cells — the body's natural "assassin" cells — from reaching and killing the cancer.

The team identified a specific molecule called LOXL2 as the architect of this barrier. In tests on mice, researchers found that by blocking LOXL2, they could dissolve the collagen wall. This allowed T cells to penetrate the tumor and significantly boosted the power of immunotherapy drugs.

"Such a strategy proved successful to enhance the antitumor effects of immunotherapy in mice with various tumor metastases," said Wang Guangchuan, a co-corresponding author of the study.

The study, which began in 2022, also utilized machine learning to simplify how doctors might predict a patient's success with treatment. The team developed a model that uses just 30 characteristic genes to predict whether a metastatic site will respond to immunotherapy, achieving an accuracy rate of over 75 percent.

The researchers are now moving toward clinical evaluations to see if the results found in mice can be replicated in humans.