Abstract 434: Understanding metabolic heterogeneity of estrogen receptor positive metastatic breast tumors using spatial transcriptomics and geographic information systems approaches

Abstract

Abstract The tumor microenvironment consists of numerous cell populations showing diverse metabolic preferences, causing metabolic heterogeneity and therapy resistance. Fulvestrant, a clinically approved selective estrogen receptor (ER) degrader, is the standard of care for ER-positive breast cancer, yet resistance has been reported among metastatic breast cancer patients. In our previous studies, we showed such resistance is attributable to metabolic heterogeneity in the tumor microenvironment. Despite metabolic heterogeneity posing a common challenge in cancer research, current methods, such as histology and Magnetic Resonance Imaging, cannot explore nutrient uptake at transcript-level resolution with spatial context, making it difficult to understand the heterogeneity that affects cellular growth and metabolism. Visium Spatial Transcriptomics, a method that shows regional gene expression, is emerging as a powerful spatial analysis tool, and we integrated this method with geographic information systems to investigate the tumor microenvironment heterogeneity and highlight the importance of cell-microenvironment interactions. We developed a computational pipeline to visualize hotspots of spatially relevant genes using breast cancer metastatic samples from xenograft models and human patients. This pipeline integrates Visium tools and geospatial methods for processing transcriptome profiling data. We analyzed the degree of local spatial association based on Local Moran’s I to estimate regions with significant co-localization of single and multiple gene expressions within a tumor. We then leveraged the co-localized areas in the process of biomarker identification and revealing biological pathways. We also used unsupervised clustering of the complete transcriptome profiles to test the resemblance between identified clusters and hotspots as well as any disruption in the spatial structure of clusters in response to therapy or metastasis. Gene set enrichment analysis of these clusters revealed several glucose or lipid metabolism pathways, and mapping of the genes corresponding to each pathway showed mutually exclusive gene expression patterns in the sample. We also observed minimal co-localization of two metabolic pathways but extensive co-localization between metabolic genes and ER target genes, suggesting distinct metabolic preferences and spatial distributions of endocrine therapy resistant cell populations. Altogether, our analysis pipeline established a computational framework for characterizing the link between tumor microenvironment, local metabolism, and changes in transcriptional response to external stimuli. Our findings support the development of strategies to target different endocrine resistant regions using a combination of ER and metabolic signaling inhibitors. Citation Format: Jin Young Yoo, Sabrina Akter, Nandana Varma, Atharva Jain, Aiman Soliman, Zeynep Madak Erdogan. Understanding metabolic heterogeneity of estrogen receptor positive metastatic breast tumors using spatial transcriptomics and geographic information systems approaches [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 434.