A new study has identified a set of plasma proteins that can predict whether patients with triple-negative breast cancer (TNBC) will respond to immunotherapy. By analyzing blood samples from 195 TNBC patients, researchers discovered that proteins like ARG1, NOS3 and CD28 are strongly linked to treatment outcomes. The team also developed a predictive model, the Plasma Immuno Prediction Score (PIPscore), which achieved high accuracy in forecasting patient responses. This non-invasive approach could revolutionize personalized treatment plans, sparing non-responders from ineffective therapies and guiding clinicians toward better alternatives. The findings highlight the critical role of systemic immunity in immunotherapy success and pave the way for more precise cancer care.

Triple-negative breast cancer (TNBC) is an aggressive subtype lacking targeted therapies, making immunotherapy a promising yet unpredictable option. Current biomarkers, such as PD-L1 expression or tumor mutational burden, often fail to reliably predict treatment success due to the complexity of immune responses. Moreover, invasive tumor biopsies are impractical for frequent monitoring. Plasma proteomics, which analyzes blood-based proteins, offers a non-invasive alternative but remains underexplored in TNBC. Previous studies have linked certain plasma proteins to immune activity, but none have systematically mapped their dynamics during immunotherapy or tied them to clinical outcomes. Based on these challenges, there is a pressing need to identify reliable, non-invasive biomarkers to optimize immunotherapy for TNBC patients.

Researchers from Fudan University Shanghai Cancer Center and Shanghai Institute for Biomedical and Pharmaceutical Technologies unveiled a plasma proteomics study of 195 TNBC patients. Using high-sensitivity assays, the team tracked 92 immune-related proteins before, during, and after immunotherapy. They identified ARG1, NOS3 and CD28 as key predictors of response and developed the PIPscore, a model with 85.8% accuracy. The study integrates single-cell RNA sequencing to correlate blood-based findings with tumor microenvironment changes, offering a holistic view of immunotherapy dynamics.

Study flowchart

Part I profiled 92 plasma proteins (Olink Target 96 Immuno-Oncology panel) across pre-, on-, and post-treatment phases in a triple-negative breast cancer (TNBC) cohort predominantly receiving immunotherapy (n = 169) with a minority undergoing chemotherapy (n = 26). Part II performed differential expression analysis of plasma proteins between pre- and post-treatment timepoints, revealing dynamic changes in protein levels accompanied by pathway enrichment that demonstrated significant immune activation signatures. Part III identified key biomarkers through differential expression analysis between responders and non-responders and developed the Plasma Immuno Prediction Score (PIPscore; AUC = 0.858) for stratifying responders to neoadjuvant immunotherapy. This study established a comprehensive plasma proteomic signature for predicting immunotherapy response in TNBC, integrating dynamic treatment monitoring with robust clinical validation through a novel PIPscore model. *P < 0.05, **P < 0.01, ***P < 0.001.

The study revealed dramatic shifts in plasma protein levels post-immunotherapy, with immune-activating proteins like CXCL9 and IFN-γ rising in responders. Notably, patients achieving pathologic complete response(pCR) had higher ARG1 and CD28 but lower NOS3 levels, suggesting these proteins regulate immune activation and tumor suppression. The PIPscore, combining six proteins (e.g., ARG1, NOS3, IL-18), stratified patients into high- and low-response groups with striking precision (AUC 0.858). High PIPscores correlated with better outcomes, while low scores indicated resistance. Single-cell RNA sequencing further linked plasma proteins to tumor microenvironment changes. For example, elevated NOS3 levels associated with fewer CD8+ T cells in tumors, hinting at immunosuppressive effects. Conversely, ARG1‘s role in arginine metabolism may enhance T-cell function. The team validated findings via ELISA, confirming the reliability of their proteomic platform.A standout highlight was the PIPscore’s prognostic power: it predicted 12-month progression-free survival with 96% accuracy. This tool could streamline clinical decision-making, identifying ideal candidates for immunotherapy upfront.

“This study transforms how we approach TNBC immunotherapy,” said Dr. Yizhou Jiang, co-corresponding author. “By translating complex plasma proteomics into a practical score, we’ve bridged the gap between research and clinical utility. The PIPscore not only predicts response but also opens doors to targeting metabolic pathways like arginine deprivation to overcome resistance. These findings underscore that systemic immunity, not just the tumor microenvironment, dictates treatment success.”

The PIPscore could soon guide oncologists in selecting TNBC patients for immunotherapy, reducing unnecessary side effects and costs. Its non-invasive nature allows repeated monitoring, enabling real-time adjustments to treatment plans. Beyond TNBC, this approach might apply to other cancers where immunotherapy efficacy varies widely.

Source – Eurekalert