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

Amin Zadeh Shirazi,Amin Zadeh Shirazi,Amin Zadeh Shirazi,Eric Fornaciari,Santosh Poonnoose,Santosh Poonnoose,Guillermo A Gomez,Rebecca J Ormsby,Michael S Samuel,Michael S Samuel,Michael S Samuel,Michael S Samuel,Kaitlin G Scheer,Kaitlin G Scheer,Kaitlin G Scheer,Damon J Tumes,Damon J Tumes,Damon J Tumes,Mark D Mcdonnell,Narjes Sadat Bagherian,Mahdi Yaghoobi

doi:10.1038/s41416-021-01394-x

Abstract

BackgroundGlioblastoma 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.MethodsWe 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.ResultsWe 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.ConclusionsThis 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.

Highlights

Glioblastoma is the most frequently diagnosed and aggressive type of brain cancer, accounting for 80% of primary malignant brain tumours of the central nervous system (CNS), and 60% of all malignant brain tumours in adults.[1]
It can be seen that the labels for training deep learning models have been generated by non-experts or automated methods and the level of noise in such labels is typically higher than the labels annotated by experts
Once individual clusters were analysed, we calculated the average expression of each gene in the gene signatures of the CTpos, ITneg and CTmvpneg regions and performed hierarchical clustering to determine which cell types within the brain tumour expresses each of the genes in the different gene signatures that we identified for each region (Fig. 4c, Supplementary Figs. 5 and 6, Supplementary Tables 11–13)

Summary

Introduction

Glioblastoma is the most frequently diagnosed and aggressive type of brain cancer, accounting for 80% of primary malignant brain tumours of the central nervous system (CNS), and 60% of all malignant brain tumours in adults.[1]. 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. 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

Methods

Results

Conclusion