Abstract Breast cancer is a leading cause of cancer-related death among Canadian women. Advances in treatment have underscored the importance of tumor heterogeneity, and the diverse cellular composition of the tumor microenvironment across molecular subtypes. The evolution of single cell imaging technologies now allows researchers to incorporate spatial context when evaluating these complex cellular interactions, allowing cellular and sub-cellular resolution across the proteome and transcriptome. In this study, a 100-core tissue microarray (TMA) from 25 breast cancer tissue blocks spanning molecular subtypes was utilized. A single 5-micron section was subjected to two consecutive rounds of staining for CosMx SMI proteomic profiling using the 64-plex Immuno-Oncology assay followed by transcriptomic profiling using the 6K RNA assay. In total, 102 fields of view (FOVs) were profiled following the CosMx Multiomic Assay protocol (MAN-10201-04). After quality control of the protein assay, we profiled 175, 139 cells identifying a mean of sixty-seven unique proteins per cell. With subsequent staining for the 6K RNA assay, 120, 797 cells were profiled, using the overlaid FOV map for downstream data stitching, identifying a total of 6, 176 genes, a mean of 365 unique genes per cell, and a mean of 556 transcripts per cell. To check concordance of cell typing, we identified HER2 as a consistent target between the clinically reported data, inclusion in the protein panel and inclusion (ERBB2) in the RNA panel. We found all core HER2 expression measured via the CosMx SMI matched the clinical annotations, and a strong correlation between HER2 protein expression and ERBB2 gene expression across the entire flow cell (r=0. 79, p=5. 71e-23). Conversely, Ki67 displayed a weak correlation (r=0. 13, p=0. 201), highlighting the importance of evaluating both metrics to discern a full picture of biological processes. CELESTA cell typing identified various structural cells expected to be found in breast tissue within the proteomic profiling (adipose, endothelial, epithelial, and vascular smooth muscle cells), along with various immune populations including B cells, CD4+, CD8+ T-cells, dendritic cells, natural killer cells, neutrophils, macrophages, and fibroblasts. These findings were corroborated in the transcriptomic data, with InSituType supervised clustering (BreastCancerWu. RData) revealing populations of cancer associated fibroblasts (CAFs), tumor associated macrophages (TAMs), tumor infiltrating lymphocytes (TILs), both naïve and memory B cells, along with basal, HER2+, Luminal A and Luminal B epithelial cell populations. In ongoing work, we continue to integrate proteomic and transcriptomic data to further investigate unknown cell populations found across the TMA and categorize fundamental differences in tumor epithelium interactions across molecular subtypes of breast cancer. Citation Format: Megan Hopkins, Oliver De Sa, Dan Dion, Vida Talebian, Jane Bayani, Melanie Spears. In situ spatial multiomic profiling of breast cancer tissue utilizing the CosMx Spatial Molecular Imager abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts) ; 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86 (7 Suppl): Abstract nr 3966.
Hopkins et al. (2026) studied this question.