Radiogenomics analysis reveals the associations of dynamic contrast-enhanced–MRI features with gene expression characteristics, PAM50 subtypes, and prognosis of breast cancer

Abstract

BackgroundTo investigate reliable associations between dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) features and gene expression characteristics in breast cancer (BC) and to develop and validate classifiers for predicting PAM50 subtypes and prognosis from DCE-MRI non-invasively.MethodsTwo radiogenomics cohorts with paired DCE-MRI and RNA-sequencing (RNA-seq) data were collected from local and public databases and divided into discovery (n = 174) and validation cohorts (n = 72). Six external datasets (n = 1,443) were used for prognostic validation. Spatial–temporal features of DCE-MRI were extracted, normalized properly, and associated with gene expression to identify the imaging features that can indicate subtypes and prognosis.ResultsExpression of genes including RBP4, MYBL2, and LINC00993 correlated significantly with DCE-MRI features (q-value < 0.05). Importantly, genes in the cell cycle pathway exhibited a significant association with imaging features (p-value < 0.001). With eight imaging-associated genes (CHEK1, TTK, CDC45, BUB1B, PLK1, E2F1, CDC20, and CDC25A), we developed a radiogenomics prognostic signature that can distinguish BC outcomes in multiple datasets well. High expression of the signature indicated a poor prognosis (p-values < 0.01). Based on DCE-MRI features, we established classifiers to predict BC clinical receptors, PAM50 subtypes, and prognostic gene sets. The imaging-based machine learning classifiers performed well in the independent dataset (areas under the receiver operating characteristic curve (AUCs) of 0.8361, 0.809, 0.7742, and 0.7277 for estrogen receptor (ER), human epidermal growth factor receptor 2 (HER2)-enriched, basal-like, and obtained radiogenomics signature). Furthermore, we developed a prognostic model directly using DCE-MRI features (p-value < 0.0001).ConclusionsOur results identified the DCE-MRI features that are robust and associated with the gene expression in BC and displayed the possibility of using the features to predict clinical receptors and PAM50 subtypes and to indicate BC prognosis.

Multi-View Radiomics Feature Fusion Reveals Distinct Immuno-Oncological Characteristics and Clinical Prognoses in Hepatocellular Carcinoma

“…Su et al developed a radiomics model for tumor-infiltrating lymphocyte status based on GLCM and GLSZM texture features [ 40 ]. Ming et al also revealed that cell cycle pathway exhibited significant associations with SurfaceVolumeRatio (PRF9 and PRF10 in our study) [ 75 , 76 ]. Hence, these results demonstrated the credibility of links between immuno-oncological characteristics and imaging features.…”

Unsupervised Analysis Based on DCE-MRI Radiomics Features Revealed Three Novel Breast Cancer Subtypes with Distinct Clinical Outcomes and Biological Characteristics

“…In the clinic, various imaging techniques are applied to diagnose BC, such as ultrasound, mammography, and MRI; among these, dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) data are widely used in BC radiomics or radiogenomics studies for their advantages in high quality and three-dimension resolution. However, most studies focus on uncovering the relationship between medical imaging features and the molecular or clinical characteristics of the disease, and few have analyzed the disease heterogeneity in imaging phenotypes from an independent perspective [ 19 , 20 , 21 , 22 , 23 , 24 , 25 ].…”

Lymph node metastasis prediction and biological pathway associations underlying DCE-MRI deep learning radiomics in invasive breast cancer

“…The relationship between image phenotypes and biological pathways has been shown in numerous studies, suggesting the possibility of observing cancer-related pathways in a non-invasive manner by DCE-MRI. For example, prognostic radiomic phenotypes of glioblastoma cells could be classified into four types based on the potential pathways of radiomic phenotypes, reflecting key biological processes related to immune regulation, tumor proliferation, therapeutic response and cellular functions, all of which affect patient survival outcomes [ 30 ]; genes in the cell cycle pathway exhibited significant associations with MRI imaging phenotypes [ 31 ]. The inhomogeneous enhancement phenotypes of tumor-adjacent parenchyma in MR imaging were associated with tumor necrosis signaling pathways [ 32 ].…”