Abstract 5079: Spatial transcriptomics analysis of early-stage ER+/HER2- breast cancer reveals unique prognostic immune-fibroblastic subtypes

Introduction: ER+ breast cancers have the best prognosis among all known breast cancer subtypes. Nonetheless, a substantial proportion of these patients, approximately 41%, experience late recurrence after at least 10 years after diagnosis. Numerous studies have highlighted the pivotal role of the tumor microenvironment (TME) in ER- breast cancer prognosis. However, this has not been thoroughly explored in the ER+ breast cancer subtype. In this study, we investigated immune and stromal cell diversity in the tumor microenvironment in hormone-dependent breast cancers and their association with survival. Methods: Using NanoString GeoMx Digital Spatial Profiling (DSP) with the Whole Transcriptome Atlas (WTA), we conducted spatial transcriptomics profiling on formalin-fixed, paraffin-embedded tissue samples (FFPE). Regions of interest (ROIs) were selected from a cohort of 410 early-stage ER+/HER2- patients’ tissue microarray (TMA). Guided by Pan-cytokeratin immunostaining (panCK), ROIs were segmented into neoplastic epithelial cells (panCK+) and TME (panCK-). Cell deconvolution was performed to elucidate immune and stromal cell abundances, using Nanostring’s Spatialdecon and CIBERSORT tools. Results: Cell abundances dichotomized into high and low groups were assessed for their association with survival. While the overall lymphocyte abundance and its subpopulations (TCD4 and TCD8 abundances) were not significantly associated with survival, they were associated with increasing Oncotype risk score. On the other hand, high fibroblast abundance was significantly associated with decreasing Oncotype risk score and a better survival outcome as a categorical and continuous variable. Based on an unsupervised clustering of patients by their immune and stromal cell proportions two distinct subgroups of patients were identified; lymphocyte-rich and lymphocyte-deficient. The lymphocyte-deficient group was characterized by high infiltration of fibroblasts and endothelial cells. Survival analysis in the subgroups yielded a significant result with the lymphocyte-rich group being associated with poorer outcomes (HR = 1.87, p = 0.04). Pathway analyses of differentially expressed genes in the lymphocyte-rich subgroup showed increased activity in immune activation, while the lymphocyte-deficient group was associated with increased activity in the extracellular matrix organization and angiogenesis. Conclusion: While this finding challenges the conventional understanding of the role of lymphocytes and fibroblasts in cancer research, this complexity throws light on the heterogeneity of the tumor microenvironment in ER+ breast cancers, and can be harnessed to improve risk stratification, for more accurate, individualized therapies. Citation Format: Hannah N. Nyarko, Zak Kinsella, Daria Kalinska-Lysiak, Niamh M.C. Connoly, Jochen Prehn, John Crown, Catherine Kelly, William Gallagher, Darran O'Connor. Spatial transcriptomics analysis of early-stage ER+/HER2- breast cancer reveals unique prognostic immune-fibroblastic subtypes [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 5079.