Abstract 2711: Deciphering transcriptional activity of the tumor microenvironment for robust stratification of chemotherapy response in triple-negative breast cancer

Abstract Background: Triple-negative breast cancer (TNBC), defined by the absence of estrogen receptor, progesterone receptor, and HER2 expression, represents an aggressive subtype with poor prognosis due to limited therapeutic options beyond chemotherapy. Current molecular subtyping methods fail to predict outcomes despite known tumor-immune-stromal interactions. The tumor microenvironment (TME) influences treatment resistance, yet capturing its complex dynamics remains challenging. TNBC demonstrates distinct features between European and Asian populations, suggesting ethnicity-specific disease mechanisms. Development of biomarkers capable of delineating TME features holds promise for improving TNBC treatment outcomes across ethnic groups. Methods: We applied our previously proposed pan-cancer biomarker TmS (tumor-specific total mRNA expression, Cao et al. Nature Biotechnology 2022) to 575 chemotherapy-treated TNBC patients from four cohorts (The Cancer Genome Atlas [TCGA], Molecular Taxonomy of Breast Cancer International Consortium [METABRIC], Sweden Cancerome Analysis Network-Breast [SCAN-B], and Fudan University Shanghai Cancer Center [FUSCC]). TmS analysis uses integrative deconvolution of matched RNA/DNA sequencing data to measure per cell per haploid total mRNA ratio for tumor versus non-tumor cells. Using recursive partitioning on survival outcomes, patients were stratified into high/low TmS groups. We performed survival analyses, gene set enrichment analysis, cell type deconvolution, and computational histology on H&E images. Results: Across all cohorts, low-TmS patients (n=325) presented significantly faster progression of TNBC (log-rank test P value < 0.05, hazard ratio = 0.1-0.53). The high vs. low TmS dichotomization represents a novel independent biomarker distinct from Lehman's TNBCTypes. TmS stratification is independent of clinical markers including BRCA mutation status, tumor mutation burden and Ki67. Mechanistically, high-TmS tumors (n=250) showed distinct TME features with enhanced immune activation in the Western cohorts but not in the Asian cohort, while low-TmS tumors demonstrated immune exclusion with extra-cellular matrix (ECM) enrichment. These molecular distinctions were validated by computational histology showing increased stromal barriers in low-TmS tumors (Kruskal test, adjusted P value < 0.05), suggesting potential ECM-targeted therapeutic strategies. Conclusions: TmS represents a clinically actionable biomarker capturing TME characteristics in TNBC. Its ability to summarize distinct TME states across diverse populations is ideal for patient stratification. The identification of ethnic-specific TME patterns and ECM-related targets in low-TmS tumors offers new therapeutic opportunities for chemotherapy-resistant TNBC. Citation Format: Yaoyi Dai, Xiaoxi Pan, Shaolong Cao, Shuai Guo, Shuangxi Ji, Matthew Montierth, Yujie Jiang, Leming Shi, Gloria V. Echeverria, Lucy Yates, Johan Staaf, Bora Lim, Yinyin Yuan, Wenyi Wang. Deciphering transcriptional activity of the tumor microenvironment for robust stratification of chemotherapy response in triple-negative breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2711.