Spatial Transcriptomics Profiling of the Tumour Microenvironment of Esophageal Adenocarcinoma

Background Despite major improvements in therapies over the last decade, esophageal adenocarcinoma (OAC) is characterized by high heterogeneity, poor treatment response and low survival rates. Chemotherapy with or without radiotherapy followed by esophagectomy remains the standard curative treatment. Including the tumor microenvironment (TME) and immune infiltrates in treatment decisions can lead to improved treatments. The spatial organization and interactions of tumor subpopulations with the TME are contributing to cancer proliferation, invasion, and immune escape. The TME of OAC is largely immunosuppressed and reflects treatment resistance as patients with inflamed TME have better outcomes. Methods We performed spatial transcriptomics on 103 samples from 34 patients, comprising 10 Barrett’s esophagus, 66 pre-treatment OAC and 27 post-treatment OAC samples. All samples were preserved in formalin-fixed paraffin-embedded (FFPE) blocks before being processed with the Visium Spatial Gene Expression kit (10X Genomics). H&E slides were annotated by our study pathologist to support pathology-guided clustering. Data analysis was performed using Seurat R pipeline, which included quality check, normalization, integration, cell type annotation and clustering of spots. Differential gene expression analyses were performed carried out across clusters and clinical groups to identify dysregulated pathways and characterize identified clusters. Results A total of 49,466 spots were phenotyped, revealing the presence of diverse cell populations, including cancer cells, stromal cells, and immune cells. We identified distinct functional clusters that correlated with pathology annotations and were characterized by specific gene expression profiles and activated pathways. We identified high intra- and inter- tumor heterogeneity, as well as different levels of immune infiltration which may correlate with patient outcomes. Comparing matching pre- and post-treatment samples revealed further differences that could be related to treatment pressure and tumor evolution. Conclusion Our study is the largest spatial profiling of OAC conducted to date. It integrates spatial transcriptomics with tissue morphology annotations of OAC and clinical data to investigate the TME landscape in relation to treatment and patient outcomes. Advances in spatial technologies offer new perspectives on the ecosystem and landscape of the OAC TME. The design of our dataset, which tracks patients across different disease stages and treatment phases, provides a unique opportunity to deepen our understanding of OAC biology and contribute to the progress of precision oncology.