Abstract 2078: Single-cell spatial transcriptomics in colon adenocarcinoma reveals tumor heterogeneity and immune microenvironmental shifts

Abstract Colon adenocarcinoma (CRC) features complex molecular changes and a dynamic remodeling of the tumor microenvironment (TME). Understanding spatial organization of gene expression in tumor and adjacent normal tissues is crucial for insights into tumorigenesis, immune modulation, and therapeutic resistance. Traditional methods like single-cell RNA sequencing (scRNA-seq) provide cellular insights but lack spatial context. Spatial transcriptomics addresses this gap, offering high-resolution, whole transcriptome analysis within tissue architecture. This study applies spatial transcriptomics to CRC, correlating molecular data with histopathological features to explore tumor heterogeneity, immune dynamics, and therapeutic targets. FFPE tissue sections (5 µm) from colon adenocarcinoma and adjacent normal tissues were analyzed using the CosMx® Spatial Molecular Imager (SMI) to create a spatially resolved whole transcriptome, single-cell map. The sections were stained with H&E for histopathological context, and spatial gene expression data was aligned with H&E-stained sections using custom software. Additionally, scRNA-seq was performed on a thicker curl (25 µm) from the same tissue block for comparative analysis. Spatial transcriptomic analysis revealed distinct gene expression patterns between tumor and normal regions, highlighting tumor heterogeneity with subpopulations of cancer cells showing varying differentiation states. Immune cell infiltration varied spatially, indicating diverse immune responses across the tumor. While both platforms identified similar transcriptional profiles, the CosMx Whole Transcriptome (WTX) panel preserved tissue architecture, allowing exploration of gene expression in its native context. CosMx WTX showed a more accurate cell composition ratio than scRNA-seq, which may introduce bias due to tissue dissociation. CosMx WTX also captured rare cell populations, such as tumor-infiltrating lymphocytes, often underrepresented in scRNA-seq due to sample size limitations. The spatial context enabled pathway activity analysis and ligand-receptor interaction mapping, providing deeper insights into tumor biology. This study highlights the power of spatial transcriptomics in colon adenocarcinoma, offering insights into tumor heterogeneity, immune infiltration, and the TME. Integrating spatial gene expression with histopathological features enhances understanding of tumor progression and immune modulation. The comparison with scRNA-seq underscores the unique benefits of spatially resolved profiling, presenting a more complete picture of cellular functions within their native tissue architecture. These findings emphasize the potential of spatial transcriptomics to advance cancer research, guide therapeutic strategies, and reveal novel insights into tumor biology. Citation Format: Margaret Hoang, Yi Cui, Shanshen He, Michael Patrick, Megan Vandenberg, Stefan Rogers, Evelyn Metzger, Haiyan Zhai, Martin Shelton, Michael Rhodes, Joseph Beechem. Single-cell spatial transcriptomics in colon adenocarcinoma reveals tumor heterogeneity and immune microenvironmental shifts [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 2078.