A deep convolutional neural network for segmentation of whole-slide pathology images identifies novel tumour cell-perivascular niche interactions that are associated with poor survival in glioblastoma

• Published in: British journal of cancer 2021-08, Vol.125 (3), p.337-350
• Main Authors: Zadeh Shirazi, Amin, McDonnell, Mark D., Fornaciari, Eric, Bagherian, Narjes Sadat, Scheer, Kaitlin G., Samuel, Michael S., Yaghoobi, Mahdi, Ormsby, Rebecca J., Poonnoose, Santosh, Tumes, Damon J., Gomez, Guillermo A.
• Format: Article
• Language: English
• Subjects: 631/114/1305; 631/67/1922; 631/67/327; Biomedical and Life Sciences; Biomedicine; Brain cancer; Brain Neoplasms - diagnostic imaging; Brain Neoplasms - genetics; Cancer Research; Correlation analysis; Deep Learning; Drug Resistance; Epidemiology; Gene Expression Profiling - methods; Gene Expression Regulation, Neoplastic; Gene Regulatory Networks; Genomes; Glioblastoma; Glioblastoma - diagnostic imaging; Glioblastoma - genetics; Humans; Image processing; Microenvironments; Microglia; Microvasculature; Molecular Medicine; Monocytes; Necrosis; Neural networks; Neural Networks, Computer; Oncology; Pericytes; Radiographic Image Interpretation, Computer-Assisted - methods; Segmentation; Single-Cell Analysis; Stem Cell Niche; Stroma; Survival Analysis; Tumor Microenvironment; Tumors
• ISSN: 0007-0920; 1532-1827
• DOI: 10.1038/s41416-021-01394-x

Background Glioblastoma is the most aggressive type of brain cancer with high-levels of intra- and inter-tumour heterogeneity that contribute to its rapid growth and invasion within the brain. However, a spatial characterisation of gene signatures and the cell types expressing these in different tumour locations is still lacking. Methods We have used a deep convolutional neural network (DCNN) as a semantic segmentation model to segment seven different tumour regions including leading edge (LE), infiltrating tumour (IT), cellular tumour (CT), cellular tumour microvascular proliferation (CTmvp), cellular tumour pseudopalisading region around necrosis (CTpan), cellular tumour perinecrotic zones (CTpnz) and cellular tumour necrosis (CTne) in digitised glioblastoma histopathological slides from The Cancer Genome Atlas (TCGA). Correlation analysis between segmentation results from tumour images together with matched RNA expression data was performed to identify genetic signatures that are specific to different tumour regions. Results We found that spatially resolved gene signatures were strongly correlated with survival in patients with defined genetic mutations. Further in silico cell ontology analysis along with single-cell RNA sequencing data from resected glioblastoma tissue samples showed that these tumour regions had different gene signatures, whose expression was driven by different cell types in the regional tumour microenvironment. Our results further pointed to a key role for interactions between microglia/pericytes/monocytes and tumour cells that occur in the IT and CTmvp regions, which may contribute to poor patient survival. Conclusions This work identified key histopathological features that correlate with patient survival and detected spatially associated genetic signatures that contribute to tumour-stroma interactions and which should be investigated as new targets in glioblastoma. The source codes and datasets used are available in GitHub: https://github.com/amin20/GBM_WSSM .