Abstract:e16249 Background: Despite the epidemiological association between cholangiocarcinoma (CCA) and HBV infection, little is known about the relevant oncogenic biomarkers for advancing cancer research and improving future patient therapy. Radiomics-based non-invasive biomarkers are promising to facilitate the translation of therapeutically related molecular subtypes for treatment allocation of patients with HBV-infected CCA. Methods: In this study, we included 55 CCA patients with whole-genome multi-omics data, proteomic data with matched magnetic resonance imaging (MRI) data from institute 1. A total of 431 patients with histopathological confirmed CCA (143 HBV-positive and 288 HBV-negative patients) were retrospectively enrolled from three institutes, and patients were allocated to the training (n = 302) and validation (n = 129) cohorts from different institutes or time period (7:3). Utilizing linear regression and gene set enrichment analysis, we identified statistically significant associations between radiomic imaging and various genomic features in HBV-positive or HBV-negative CCA patients. Additionally, we employed a random forest classifier based on radiomic features to predict the status of HBV infection, HBV-related gene mutations and protein expression profiling, and tested on an internal and an external CCA cohort (n = 53). Results: The analysis revealed significant and widespread associations between genomic features (somatic mutations, transcriptional activities and copy number variations) and radiomic features characterizing HBV infection status. Our results show that quantitative image features are capable of distinguishing several molecular phenotypes. We obtained significant predictive performance for HBV+ (AUC = 0.73), CA-199(AUC = 0.77) and NeuroInvasion (AUC = 0.71), activation of ESRP1(AUC = 0.79) and mutation of HNRNPUL2/MUC4 (AUC = 0.73). We externally validated the HBV prediction model (AUC = 0.76) on the internal and external cohort. The sensitivity, specificity, and accuracy were all > 70%. The radiomic signature succeeded to stratify patients into distinct survival outcomes according to HBV status, and remained an independent prognostic factor in multivariable analysis (DFS, P < 0.0001; OS, P = 0.029). Integrating the radiomic signature with clinicopathological-radiographic factors further improved prediction performance. A nomogram incorporating significant pathological predictors (HBV, CA-125, CA-199, Neuro invasion, peritoneal invasion, pT stage, pN stage, and differentiation, grade) and radiomics signatures displayed good prediction performance in estimating OS with AUC values of 0.895, 0.935, and 0.907 at 1, 3, and 5 years, respectively. Conclusions: This deep learning-driven combined radiomics model may provide as a useful tool to preoperatively predict HBV infection and improve therapeutic management of ICC patients.

Associations between radiomic phenotypes and genomic mechanisms in HBV-associated cholangiocarcinoma.

Radiomics Analysis of Contrast-Enhanced CT for the Preoperative Prediction of Microvascular Invasion in Mass-Forming Intrahepatic Cholangiocarcinoma

CT-Based Radiomics for the Recurrence Prediction of Hepatocellular Carcinoma after Surgical Resection

An MRI-Based Radiomic Model for Individualized Prediction of Hepatocellular Carcinoma in Patients With Hepatitis B Virus-Related Cirrhosis

A Multidimensional Nomogram Combining Imaging Features and Clinical Factors to Predict the Invasiveness and Metastasis of Combined Hepatocellular Cholangiocarcinoma

Preoperative Radiomics Analysis of Contrast-Enhanced CT for Microvascular Invasion and Prognosis Stratification in Hepatocellular Carcinoma

A nomogram based on radiomics to predict microvascular invasion and overall survival in intrahepatic cholangiocarcinoma

Integrating Pathomics and Radiomics for Predicting Survival among Patients with Intrahepatic Cholangiocarcinoma Undergoing Surgical Resection.

Predicting Microvascular Invasion in Hepatocellular Carcinoma Using CT-based Radiomics Model.

Classification of Hepatocellular Carcinoma and Intrahepatic Cholangiocarcinoma Based on Radiomic Analysis.

Radiomics Analysis on Multiphase Contrast-Enhanced CT: A Survival Prediction Tool in Patients with Hepatocellular Carcinoma Undergoing Transarterial Chemoembolization

Assessing microvascular invasion in HBV-related hepatocellular carcinoma: an online interactive nomogram integrating inflammatory markers, radiomics, and convolutional neural networks

Identification and Validation of a Predictive Immune-related Genes Signature for the Prognosis of Cholangiocarcinoma

Association Between MRI‐Based Radiomics Features and Regional Lymph Node Metastasis in Intrahepatic Cholangiocarcinoma and Its Clinical Outcome

A Prognostic Signature Based on Immune-Related Genes Associated with Tumor Microenvironment for Cholangiocarcinoma

Comprehensive radiomics nomogram for predicting survival of patients with combined hepatocellular carcinoma and cholangiocarcinoma

Development and Validation of Combined Ki67 Status Prediction Model for Intrahepatic Cholangiocarcinoma Based on Clinicoradiological Features and MRI Radiomics

A CT-based radiomics approach to predict intra-tumoral tertiary lymphoid structures and recurrence of intrahepatic cholangiocarcinoma

Combined Hepatocellular Carcinoma-Cholangiocarcinoma: MRI Features Correlated with Tumor Biomarkers and Prognosis

Evaluating Microvascular Invasion in Hepatitis B Virus-Related Hepatocellular Carcinoma Based on Contrast-Enhanced Computed Tomography Radiomics and Clinicoradiological Factors

A Radiomics Approach to Predict Lymph Node Metastasis and Clinical Outcome of Intrahepatic Cholangiocarcinoma.