Spatial whole transcriptome profiling of the tumor microenvironment in FFPE prostate carcinoma using the Visium platform

Abstract Despite years of prostate cancer research, balancing the risks of treatment complications and tumor progression remains a challenge. The inability to dissect tumor microenvironments (TME) and immune compartments contributes to a knowledge gap. Spatially resolved molecular profiling allows understanding these complexities for potential personalized treatments. We used 10× Genomics Visium Spatial Gene Expression for FFPE tissue for spatial whole transcriptome analysis of two prostate cancer sample TMEs. The tissues were processed over~5,000 molecularly barcoded, spatially encoded capture probes in spots for spatially resolved transcriptional readout. We resolved whole transcriptome tumorigenic profiles in sections of normal, stage III adenocarcinoma, and stage IV acinar cell carcinoma FFPE human prostate tissues. Computational clustering identified spatial gene expression patterns that aligned with pathologist annotations. The data revealed spatial disorganization of basal and luminal cells in tumor samples. T lymphocytes were dispersed throughout the whole tissue in the adenocarcinoma, while plasma B cells were in the peritumoral region impacting patient prognosis. Computational methods inferred copy number variation, identifying aneuploidy regions and specific loci driving the genomic profile of the cancerous regions. We demonstrated that spatial whole transcriptome analysis resolves FFPE prostate samples. Our data rapidly confirms known cell type patterns and tumor region-specific gene expression while enhancing understanding of the TME for drug target or biomarker discovery for tailored therapies and patient stratification.