Poster #057, University Of California, San Francisco
Spatial Gene Expression Analysis of African American Prostate Cancer Tissue Reveals Distinct Cell Types and Patterns through Xenium
Mentors: Verima Pereira, PhD and Franklin Huang, MD, PhD
Prostate cancer is characterized by marked cellular and spatial heterogeneity that contributes to tumor progression, therapeutic resistance, and clinical outcomes. Characterizing the tumor microenvironment at high spatial resolution is essential for understanding interactions between malignant, stromal, and immune cells. Spatial transcriptomic profiling of prostate tumor tissue microarrays (TMAs) provides a powerful platform for investigating these complex cellular ecosystems while preserving tissue architecture. Prostate tumor tissue microarrays (TMAs) were profiled using spatial transcriptomics. Data were analyzed with the Seurat workflow, including quality control, normalization, unsupervised clustering, and UMAP visualization. Cell types were annotated using canonical marker genes, while malignant populations were classified based on tumor-associated gene signatures (AGR2, AMACR, ERG, ETV4, MYC, PCAT14, IFN, and hypoxia). Cellular composition and spatial distribution were compared across 30 TMA cores to assess tumor heterogeneity. UMAP analysis identified 20 distinct cellular clusters, clearly separating malignant and non-malignant populations. Annotation using canonical cell markers and tumor-associated gene signatures revealed diverse epithelial, stromal, immune, and malignant cell populations. Comparison of 30 TMA cores demonstrated marked spatial heterogeneity, with individual cores enriched for distinct tumor-associated transcriptional states (AGR2, AMACR, ERG, ETV4, MYC, PCAT14, IFN, and hypoxia) and varying proportions of fibroblasts, endothelial cells, and immune cell populations, highlighting diverse tumor microenvironmental niches. Spatial transcriptomic profiling of prostate tumor TMAs reveals extensive cellular and molecular heterogeneity across individual tissue cores. Integrating cell-type annotation with spatial composition analysis provides a comprehensive map of the prostate tumor microenvironment and identifies distinct malignant, stromal, and immune niches. These findings demonstrate the utility of TMA-based spatial transcriptomics for advancing our understanding of prostate cancer biology, facilitating biomarker discovery, and supporting the development of precision medicine strategies.