Spatial transcriptomics of glioblastoma defines biologically and clinically significant reprogramming patterns across unique spatial microenvironments

Glioblastoma (GBM) is the most common and lethal primary malignant tumor of the central nervous system. Advances in therapy are hindered by the complex intratumoral heterogeneity of GBM, where distinct malignant and non-malignant cellular states and interactions exist in spatially defined niches of the tumor microenvironment (TME), shaping both tumor behavior and treatment response. In this work, we define GBM biological reprogramming, TME recomposition, and cell-cell interactions in relation to spatially well defined Ivy Glioblastoma Atlas Project regions. Further, we apply a novel spatially informed approach to integrate biological reprogramming, as identified through gene co-expression network modules, with specific changes in cell-cell communications. Our results validate prior findings and offer novel insights into the spatial patterning of transcription factor regulation, cellular interactions, and biological pathway activity, in addition to informing rational combination therapies targeting spatial niche specific vulnerabilities. In addition, our work contributes a novel Visium spatial transcriptomics dataset of 14 samples for use by the GBM research community.