AI and Cancer Research ── Liver ── 2026-10-26
Liver cancer: treatment, trials and AI
- Abstract
PURPOSE: To develop a Magnetic resonance imaging (MRI) -based radiomics-clinical model for noninvasive prediction of fibroblast activation protein (FAP) expression in hepatocellular carcinoma (HCC), and to explore its association with tumor microenvironment (TME) characteristics and potential immunotherapy response. PATIENTS AND METHODS: This retrospective study included 104 patients with HCC from the Fourth Hospital of Hebei Medical University, 31 external MRI cases from The Cancer Imaging Archive (TCIA), and 363 transcriptomic profiles from The Cancer Genome Atlas (TCGA). Immunohistochemistry (IHC) assessed FAP expression. Tumor regions were manually segmented, and radiomics features were extracted using PyRadiomics. Multiple machine learning classifiers were used to construct and validate models for predicting FAP expression. Logistic regression was selected for its balanced performance between the training and validation cohorts, with the smallest area under the curve (AUC) difference among the five machine-learning algorithms. A radiomics-clinical combined model was further developed by integrating the radscore with clinical indicators. Radiogenomic analysis explored radscore-related transcriptomic features and immune infiltration characteristics. RESULTS: In the validation cohort, the AUC values of the intratumoral, peritumoral, and intratumoral + peritumoral radiomics models were 0.757, 0.502, and 0.784, respectively. The intratumoral + peritumoral model also achieved a sensitivity of 0.692 and a specificity of 0.788, indicating the best overall discriminatory performance among the three models. The radiomics-clinical combined model integrating the systemic immune-inflammation index (SII), prognostic nutritional index (PNI), and radscore achieved an AUC of 0.816 in the validation cohort. Radscore-related transcriptomic modules identified 10 hub genes for risk stratification. Patients in the high-risk group exhibited a more pronounced immunosuppressive TME and poorer predicted benefit from immunotherapy. CONCLUSION: The MRI-based radiomics-clinical model effectively predicted FAP expression in HCC. Radscore reflected FAP-related immunosuppressive TME remodeling and may provide insights into potential immunotherapy response stratification, which requires further validation in independent immunotherapy-treated cohorts.
Journal IF-equivalent: 2.9 (OpenAlex 2-year mean citedness, value as of 2026-10-02, retrieved 2026-10-03; not the official Clarivate IF)Reference: Gao R, Han G, Wang Q, Zhang M, Yin F. MRI-Based Radiomics Predicts FAP Expression and Immunosuppressive Tumor Microenvironment in Hepatocellular Carcinoma. J Hepatocell Carcinoma. 2026 Oct;Volume 13:1-18. doi:10.2147/jhc.s641948. PMID: 42924263.Checked: Abstract only - Abstract
BACKGROUND: Although combining systemic and locoregional therapies has improved outcomes for hepatocellular carcinoma (HCC), identifying appropriate candidates remains challenging because no reliable tool predicts response. METHODS: This retrospective multicenter study enrolled 334 HCC patients who received first-line locoregional-systemic therapy from nine centers in China. Following automated tumor segmentation using nnU-Net on pretreatment contrast-enhanced magnetic resonance imaging, predefined radiomics features were computationally extracted, and deep learning features were obtained via a ResNet convolutional neural network. Three types of predictive models for tumor response (radiomics, deep learning, and combined) were developed and validated on the training cohort ( n = 213), internal validation cohort (IVC, n = 53), and external validation cohort (EVC, n = 68). Performance was assessed using area under the receiver operating characteristic curve (AUC) and net reclassification improvement (NRI). Associations of model with progression-free survival (PFS) and overall survival (OS) were evaluated. RESULTS: Among the three models (multi-sequence radiomics, multi-sequence deep learning, and combined), the combined model showed the best predictive performance in both the training (AUC: 0.997, 95% CI: 0.994-1.000) and validation cohorts (AUC of IVC 0.900, 95% CI: 0.806-0.994; AUC of EVC: 0.871, 95% CI: 0.790-0.951). The combined model demonstrated significantly improved predictive performance compared with the multi-sequence radiomic model (training NRI = 43.9%, p p = 0.038; EVC NRI = 3.2%, p = 0.806) and multi-sequence deep learning model (training NRI = 80.8%, IVC NRI = 72.7%, EVC NRI = 50.4%; all p < 0.05). Specifically, the combined model was significantly associated with both PFS and OS. CONCLUSIONS: The combined radiomics-deep learning model outperformed both single-modality models in predicting objective response and stratifying survival outcomes in unresectable HCC patients receiving combined systemic and locoregional therapy.
Journal IF-equivalent: 5.3 (OpenAlex 2-year mean citedness, value as of 2026-10-24, retrieved 2026-10-26; not the official Clarivate IF)Reference: Wang LN, Zhou SQ, Li F, Wu LF, Xu L, Wang K, et al. Integrating Magnetic Resonance Imaging-Based Deep Learning and Radiomics to Predict Treatment Response in Hepatocellular Carcinoma Receiving Combined Systemic and Locoregional Therapies: A Multicenter Study. Liver Cancer. 2026 Sep 7 [Epub ahead of print]. doi:10.1159/lic/admag009. PMID: 42924067.Checked: Abstract only