Graph Neural Network Model for Prediction of Non-Small Cell Lung Cancer Lymph Node Metastasis Using Protein-Protein Interaction Network and 18F-FDG PET/CT Radiomics

Abstract

The image texture features obtained from ¹⁸F-fluorodeoxyglucose positron emission tomography/computed tomography (¹⁸F-FDG PET/CT) images of non-small cell lung cancer (NSCLC) have revealed tumor heterogeneity. A combination of genomic data and radiomics may improve the prediction of tumor prognosis. This study aimed to predict NSCLC metastasis using a graph neural network (GNN) obtained by combining a protein-protein interaction (PPI) network based on gene expression data and image texture features. ¹⁸F-FDG PET/CT images and RNA sequencing data of 93 patients with NSCLC were acquired from The Cancer Imaging Archive. Image texture features were extracted from ¹⁸F-FDG PET/CT images and area under the curve receiver operating characteristic curve (AUC) of each image feature was calculated. Weighted gene co-expression network analysis (WGCNA) was used to construct gene modules, followed by functional enrichment analysis and identification of differentially expressed genes. The PPI of each gene module and genes belonging to metastasis-related processes were converted via a graph attention network. Images and genomic features were concatenated. The GNN model using PPI modules from WGCNA and metastasis-related functions combined with image texture features was evaluated quantitatively. Fifty-five image texture features were extracted from ¹⁸F-FDG PET/CT, and radiomic features were selected based on AUC (n = 10). Eighty-six gene modules were clustered by WGCNA. Genes (n = 19) enriched in the metastasis-related pathways were filtered using DEG analysis. The accuracy of the PPI network, derived from WGCNA modules and metastasis-related genes, improved from 0.4795 to 0.5830 (p < 2.75 × 10^-12). Integrating PPI of four metastasis-related genes with ¹⁸F-FDG PET/CT image features in a GNN model elevated its accuracy over a without image feature model to 0.8545 (95% CI = 0.8401-0.8689, p-value < 0.02). This model demonstrated significant enhancement compared to the model using PPI and ¹⁸F-FDG PET/CT derived from WGCNA (p-value < 0.02), underscoring the critical role of metastasis-related genes in prediction model. The enhanced predictive capability of the lymph node metastasis prediction GNN model for NSCLC, achieved through the integration of comprehensive image features with genomic data, demonstrates promise for clinical implementation.

Keywords: 18F-FDG PET; CT; GNN; NSCLC; protein–protein interaction; radiogenomics.

Figure 1

Weighted gene co-expression network analysis. A dendrogram illustrating 86 modules, where genes are clustered based on dissimilarity, highlighted with a color band showing results from an automatic single-block analysis (left); Associations between gene modules and lymph node metastasis traits were assessed by examining the correlations between the eigengenes of these modules and the traits. Each block in the heatmap represents a different module and modules were enriched with clustered genes. The intensity of the color in each block reflects the strength of the correlation between the module and the lymph node metastasis trait in non-small cell lung cancer (right). The intensity of the color in each block reflects the strength of the correlation between the module and the lymph node metastasis trait in non-small cell lung cancer Sky Blue 1 and Yellow Green modules were selected for functional analysis due to their significant correlations, evidenced by p-values of 0.03 and 0.047, respectively.

Figure 2

The receiver operating characteristic curve of six image features, comprising three attributes each from ¹⁸F-FDG PET and CT images, which have demonstrated the highest area under the curve receiver operating characteristic curve values. p-values were calculated based on the Hanley and McNeil standard error formula.

Figure 3

Scheme of the non-small cell lung cancer lymph node metastasis prediction.

Figure 4

Structure of the graph neural network model for protein–protein interaction module characteristic recognition.