Abstract
Abstract Intra-tumoral heterogeneity and cell-state plasticity are key drivers for the therapeutic resistance of glioblastoma (GBM). Studies based on single-cell RNA-seq and spatial transcriptomics have classified GBM cells into distinct transcriptional phenotypes. However, how the transcriptional diversity and spatial cellular organization are associated with patient prognosis remains incompletely resolved. Here, we developed a deep learning model to predict spatially resolved transcriptional programs from histology images. The model was trained on spatial transcriptomics data and validated in external testing cohorts. Applying the model to two separate patient cohorts led to the discovery of conserved relationships between tumor architecture and prognosis. Patient with poor prognosis had higher proportions of GBM cells expressing a hypoxia-induced transcriptional program. In addition, high clustering patterns of reactive astrocytes were associated with a poor prognosis. Conversely, when the reactive astrocytes were dispersed and connected to other cell types, the risk was decreased. To validate our results, we developed a separate deep learning model that used histology images to predict prognosis. Applying the model to spatial transcriptomics data discovered survival-associated regional gene expression programs. Genes related to glycoprotein metabolism and injury response were significantly upregulated in tumor cells with high aggressiveness. Our studies established a scalable approach to resolve the transcriptional heterogeneity of GBM and linked the spatial cellular architecture to clinical outcomes. Citation Format: Yuanning Zheng, Francisco Carrillo-Perez, Marija Pizurica, Olivier Gevaert. Spatial cellular architecture predicts prognosis in glioblastoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 2 (Clinical Trials and Late-Breaking Research); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(8_Suppl):Abstract nr LB272.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
