New breast cancer classification system predicts immunotherapy response success

The advent of immune checkpoint inhibitors (ICIs) has marked a paradigm shift in cancer treatment, yet a significant portion of breast cancer patients fail to respond to these therapies. The cancer-immunity cycle (CIC) is a conceptual framework that maps the step-by-step process of the anti-tumor immune response, from the release of cancer cell antigens to the killing of tumor cells by T cells. A defect in any single step can halt the entire cycle and render immunotherapy ineffective. However, most research has focused on individual steps, which fails to capture the complexity of the immune response. Due to these limitations, a more holistic and systematic approach is urgently needed to accurately assess a patient's immune status and guide treatment decisions.

Researchers from the Department of Breast Surgery at Fudan University Shanghai Cancer Center and the Department of Oncology at Shanghai Medical College, Fudan University have developed a new classification system for breast cancer based on the CIC. Published (DOI: 10.20892/j.issn.2095-3941.2025.0611)in Cancer Biology & Medicine in 2026, the study details how this novel framework can predict patient response to ICIs and identifies new therapeutic targets to overcome treatment resistance.

The team developed a "CIC score" to measure the activity of six key steps in the anti-tumor immune response. By analyzing thescoreof each step, they classified patients into three distinct CIC clusters. The first cluster (C1) was characterized as an "immune-cold" tumor with low immune infiltration, a poor prognosis, and an abundance of immunosuppressive M2 macrophages. In stark contrast, the third cluster (C3) represented an "immune-hot" tumor, showing high immune cell infiltration, active T cells, and the best response to ICI therapy.

The most unexpected finding was the second cluster (C2), an intermediate subtype with a unique defect in antigen presentation. Despite a high tumor mutational burden (TMB), which typically suggests responsiveness to immunotherapy, C2 tumors exhibited frequent human leukocyte antigen (HLA) loss of heterozygosity and an immunosuppressive tumor microenvironment (TME) enriched with dysfunctional dendritic cells (DCs) and regulatory T cells (Tregs). Multi-omic analyses revealed specific metabolic dependencies for each cluster, with C1 showing sphingolipid metabolism enrichment and C2 showing a strong dependency on serine metabolism. Notably, the enzyme PSAT1 was identified as a key metabolic regulator in C2, and its knockdown in cancer cells reduced the expression of key immunosuppressive molecules like *PD-L1* and TGFB1.

"The CIC provides a powerful framework for understanding how tumors evade the immune system," the authors said. "By building a comprehensive score that captures the efficiency of this entire cycle, we've moved beyond the simple 'hot' and 'cold' tumor paradigm to identify distinct, actionable defects. This allows us to not only predict which patients will benefit from current immunotherapies but also to see exactly where the cycle is breaking down, pointing us toward new, more targeted combination strategies to fix those breaks and improve outcomes for a wider range of patients."

This new classification system has immediate and far-reaching implications for clinical practice. It provides a robust biomarker, the CIC score, which could be used to stratify breast cancer patients, identifying those most likely to respond to ICI therapy and sparing others from unnecessary side effects. More importantly, the discovery of distinct immune-evasion mechanisms in each subtype paves the way for novel combination therapies. For patients with C1 tumors, treatments might need to focus on converting the "cold" microenvironment into a "hot" one, while for C2 patients, strategies to enhance antigen presentation, potentially by targeting PSAT1 or overcoming HLA loss, could be key.

Source:

Chinese Academy of Sciences

Journal reference: