Approximately 10-15% of all breast cancers are triple-negative (TNBC), defined by the absence of estrogen receptor (ER), progesterone receptor (PR), and HER2 (HER2) expression. TNBC represents a biologically aggressive subtype associated with worse prognosis compared with other breast cancer subtypes. In high-risk early-stage TNBC, the addition of immune checkpoint inhibitors to neoadjuvant chemotherapy significantly improves pathological complete response and overall survival, and it is now incorporated into international treatment guidelines. Despite these advances, reliable predictive biomarkers for response to neoadjuvant immunotherapy remain lacking. According to international guidelines, ER expression <1% is classified as ER-negative. However, studies indicate that tumors with low ER expression share biological characteristics and clinical outcomes more closely resembling ER-zero than ER-high disease. This project aims to redefine the clinical and biological significance of ER expression across the ER continuum spanning from ER-zero (including TNBC) to ER-high breast cancers, with the ultimate aim of improving patient selection for immunotherapy and endocrine therapy. Although population-based studies have described the clinical landscape of ER-zero and ER-low breast cancer, important knowledge gaps remain regarding the molecular mechanisms underlying ER-low and ER-intermediate tumors and their response to treatment. We will perform deep molecular profiling in a well-characterized cohort spanning ER-zero, ER-low, ER-intermediate, and ER-high breast cancers. Integrative multi-omics analyses will elucidate the genetic and epigenetic determinants of ER expression and endocrine responsiveness, refine the biological definition of ER-low and ER-intermediate tumors, and identify clinically relevant subgroups. Among the analysis we have performed spatial transcriptomic analysis of 40 breast tumors using the 10x Genomics Xenium 5K gene panel. After QC, we generated more than 6.3 million high quality single cell resolution spatial transcriptomic data. We will need the computing capacity powered by Dardel@PDC to support the integrative analysis of such large dataset, which is important to understand the molecular mechanism and spatial distribution of such ER expression in relation to histology features.