Cell graph neural networks enable the precise prediction of patient survival in gastric cancer

Abstract

Gastric cancer is one of the deadliest cancers worldwide. An accurate prognosis is essential for effective clinical assessment and treatment. Spatial patterns in the tumor microenvironment (TME) are conceptually indicative of the staging and progression of gastric cancer patients. Using spatial patterns of the TME by integrating and transforming the multiplexed immunohistochemistry (mIHC) images as Cell-Graphs, we propose a graph neural network-based approach, termed Cell−GraphSignatureorCGSignature, powered by artificial intelligence, for the digital staging of TME and precise prediction of patient survival in gastric cancer. In this study, patient survival prediction is formulated as either a binary (short-term and long-term) or ternary (short-term, medium-term, and long-term) classification task. Extensive benchmarking experiments demonstrate that the CGSignature achieves outstanding model performance, with Area Under the Receiver Operating Characteristic curve of 0.960 ± 0.01, and 0.771 ± 0.024 to 0.904 ± 0.012 for the binary- and ternary-classification, respectively. Moreover, Kaplan–Meier survival analysis indicates that the “digital grade” cancer staging produced by CGSignature provides a remarkable capability in discriminating both binary and ternary classes with statistical significance (P value < 0.0001), significantly outperforming the AJCC 8th edition Tumor Node Metastasis staging system. Using Cell-Graphs extracted from mIHC images, CGSignature improves the assessment of the link between the TME spatial patterns and patient prognosis. Our study suggests the feasibility and benefits of such an artificial intelligence-powered digital staging system in diagnostic pathology and precision oncology.

Quantitative Spatial Profiling of Immune Populations in Pancreatic Ductal Adenocarcinoma Reveals Tumor Microenvironment Heterogeneity and Prognostic Biomarkers

“…Identifying differences first and then correlating to survival may weaken the significance due to misclassification of certain patients. While different splitting strategies have been employed by previous studies for biomarker discovery (25,30,31), there is no consensus on the optimal split approach. In this study, 50% was selected as the cut-off point produced two groups of comparable size, also reducing the risk of overfitting for the downstream analysis.…”

Graph neural networks learn emergent tissue properties from spatial molecular profiles

“…The colorectal cancer dataset (Schürch et al . 7 with 140 images from 35 patients) was measured with CODEX. The dataset consists of two patient groups, one group with Crohn’s-like reaction (CLR) represented in 68 images and one group with diffuse inflammatory infiltration (DII) represented in 72 images.…”

“…Spatial information and graphs of cells extend the capability of such supervised models to not only detect cellular phenotypes that correlate with the labels, but also motifs of tissue architecture, such as cellular niches 4 . Because of their explicit representation of cells as constituent building blocks of a tissue, graph neural networks promise to be more interpretable with respect to niches than convolutional neural networks on tissue images and, indeed, have been recently successfully deployed for tumor phenotype prediction from spatial proteomics data [5][6][7] . Here, we perform a comparative ablation study over spatial features and single-cell resolution for graph neural networks that predict tumor phenotypes from spatial graphs with gene expression or categorical cell type node features.…”

Screening of normal endoscopic large bowel biopsies with artificial intelligence: a retrospective study

“…Each node typically has an associated k -dimensional feature vector x i for i ∈ V . In existing methods, an edge e i,j is constructed if the Euclidean distance between the centroids of nodes i and j is less than a certain threshold 9 14 . The distance between neighbouring node centroids is suitable for convex node entities, such as nuclei, because centroids will usually be located within the object.…”