Deep learning features from diffusion tensor imaging improve glioma stratification and identify risk groups with distinct molecular pathway activities

Abstract

Background To develop and validate a deep learning signature (DLS) from diffusion tensor imaging (DTI) for predicting overall survival in patients with infiltrative gliomas, and to investigate the biological pathways underlying the developed DLS. Methods The DLS was developed based on a deep learning cohort (n = 688). The key pathways underlying the DLS were identified on a radiogenomics cohort with paired DTI and RNA-seq data (n=78), where the prognostic value of the pathway genes was validated in public databases (TCGA, n = 663; CGGA, n = 657). Findings The DLS was associated with survival (log-rank P < 0.001) and was an independent predictor ( P < 0.001). Incorporating the DLS into existing risk system resulted in a deep learning nomogram predicting survival better than either the DLS or the clinicomolecular nomogram alone, with a better calibration and classification accuracy (net reclassification improvement 0.646, P < 0.001). Five kinds of pathways (synaptic transmission, calcium signaling, glutamate secretion, axon guidance, and glioma pathways) were significantly correlated with the DLS. Average expression value of pathway genes showed prognostic significance in our radiogenomics cohort and TCGA/CGGA cohorts (log-rank P < 0.05). Interpretation DTI-derived DLS can improve glioma stratification by identifying risk groups with dysregulated biological pathways that contributed to survival outcomes. Therapies inhibiting neuron-to-brain tumor synaptic communication may be more effective in high-risk glioma defined by DTI-derived DLS. Funding A full list of funding bodies that contributed to this study can be found in the Acknowledgements section.

Differences in the MRI Signature and ADC Values of Diffuse Midline Gliomas with H3 K27M Mutation Compared to Midline Glioblastomas

“…DWI techniques allow for the analysis of water mobility and tissue microstructure [35,36]. It can also be used to analyze the microstructure of brain tumors supporting differential diagnosis and grading [37][38][39] and to help evaluate therapy effects [40,41]. ADC values can vary within a single DMG between enhancing and non-enhancing tumor areas, maybe more frequently in cases with ATRX loss [33].…”

Artificial Intelligence Meets Whole Slide Images: Deep Learning Model Shapes an Immune-Hot Tumor and Guides Precision Therapy in Bladder Cancer

“…The Kaplan-Meier method was then applied to explore the prognostic significance of the AI score. Based on the gene expression profile, the potential biological pathways underlying WSI cluster and AI cluster were identified [ 29 ]. First, differentially expressed genes (DEGs) between every subgroup stratified by the WSI cluster were identified using DESeq2 R package.…”

Alternations and Applications of the Structural and Functional Connectome in Gliomas: A Mini-Review

“…On the other hand, numerous studies have reported that gliomas have specific molecular parameters, such as IDH-mutation ( Chen et al, 2017 ; Kesler et al, 2019 ) and MGMT promoter methylation ( Chen et al, 2017 ). To date, deep neural networks have been utilized to extract dMRI metrics like ADC and FA value, and to assist IDH, MGMT, and other molecule subtype classification and prognosis prediction of glioma patients ( Aliotta et al, 2019 ; Z. Huang et al, 2021 ; Yan et al, 2021 ).…”