Radiomics in the diagnosis and treatment of hepatocellular carcinoma

Artificial intelligence in imaging for liver disease diagnosis

Liver diseases, including hepatitis, non-alcoholic fatty liver disease (NAFLD), cirrhosis, and hepatocellular carcinoma (HCC), remain a major global health concern, with early and accurate diagnosis being essential for effective management. Imaging modalities such as ultrasound (US), computed tomography (CT), and magnetic resonance imaging (MRI) play a crucial role in non-invasive diagnosis, but their sensitivity and diagnostic accuracy can be limited. Recent advancements in artificial intelligence (AI) have improved imaging-based liver disease assessment by enhancing pattern recognition, automating fibrosis and steatosis quantification, and aiding in HCC detection. AI-driven imaging techniques have shown promise in fibrosis staging through US, CT, MRI, and elastography, reducing the reliance on invasive liver biopsy. For liver steatosis, AI-assisted imaging methods have improved sensitivity and grading consistency, while in HCC detection and characterization, AI models have enhanced lesion identification, classification, and risk stratification across imaging modalities. The growing integration of AI into liver imaging is reshaping diagnostic workflows and has the potential to improve accuracy, efficiency, and clinical decision-making. This review provides an overview of AI applications in liver imaging, focusing on their clinical utility and implications for the future of liver disease diagnosis.

Computed tomography radiomics combined with clinical parameters for hepatocellular carcinoma differentiation: a machine learning investigation

Purpose

To evaluate the performance of a combined clinical-radiomics model using multiple machine learning approaches for predicting pathological differentiation in hepatocellular carcinoma (HCC).

Material and methods

A total of 196 patients with pathologically confirmed HCC, who underwent preoperative computed tomography (CT) were retrospectively enrolled (training: n = 156; validation: n = 40). The modelling process included the folowing: (1) clinical model construction through logistic regression analysis of risk factors; (2) radiomics model development by comparing 6 machine learning classifiers; and (3) integration of optimal clinical and radiomic features into a combined model. Model performance was assessed using the area under the curve (AUC), calibration curves, and decision curve analysis (DCA). A nomogram was constructed for clinical implementation.

Results

Two clinical risk factors (BMI and CA153) were identified as independent predictors of differentiated HCC. The clinical model showed moderate performance (AUC: training = 0.705, validation = 0.658). The radiomics model demonstrated improved prediction capability (AUC: training = 0.840, validation = 0.716). The combined model achieved the best performance in differentiating HCC pathological grades (AUC: training = 0.878, validation = 0.747).

Conclusions

The integration of CT radiomics features with clinical parameters through machine learning provides a promising non-invasive approach for predicting HCC pathological differentiation. This combined model could serve as a valuable tool for preoperative treatment planning.

Habitat radiomics based on CT images to predict survival and immune status in hepatocellular carcinoma, a multi-cohort validation study

BACKGROUND AND OBJECTIVE

Though several clinicopathological features are identified as prognostic indicators, potentially prognostic radiomic models are expected to preoperatively and noninvasively predict survival for HCC. Traditional radiomic models are lacking in a consideration for intratumoral regional heterogeneity. The study aimed to establish and validate the predictive power of multiple habitat radiomic models in predicting prognosis of hepatocellular carcinoma (HCC).

METHODS

A total of 232 HCC patients were retrospectively included, including a training/validation cohort and two external testing cohorts from 4 centers. For habitat radiomics, intratumoral habitat partitioning based on CT images was first performed by using Otsu thresholding method. Second, a total of 350 habitat radiomic models were constructed to select the optimal model. Then, both ROC curve analyses and Kaplan-Meier survival curve analyses were applied to assess the predictive performances. Ultimately, an immune status profiling was conducted based on bioinformatic analyses and multiplex immunohistochemistry (mIHC) assays to reveal the potential mechanisms.

RESULTS

A total of 4 habitats were segmented, and the corresponding habitat radiomic models were constructed based on each habitat and an integration of all the four habitats. Generally, habitat radiomic models outperformed traditional radiomic models in stratifying prognosis for HCC. The habitat radiomic model based on the segmented habitat 4 involving decision tree (DT) screening and random forest (RF) classifier was identified as the optimal model with an AUCmean of 0.806. Distinct resting natural killer (NK) cell infiltrations significantly contributed to the prognosis stratification of HCC by the optimal habitat radiomic model.

CONCLUSIONS

The habitat radiomic model based on CT images was potentially predictive of overall survival for HCC, with a superiority over the traditional radiomic model. The prognostic power of the habitat radiomic model was partly attributed to the distinct immune status captured in the CT images.

Ultrasound-based Radiomics Analysis for Assessing Risk Factors Associated With Early Recurrence Following Surgical Resection of Hepatocellular Carcinoma

OBJECTIVE

The aim of this study was to explore the value of ultrasound-based radiomics analysis for early recurrence after surgical resection of hepatocellular carcinoma (HCC).

METHODS

This retrospective study included 127 patients who underwent primary surgical resection for HCC between October 2019 and November 2021. The patients were subsequently divided into training and validation sets (7:3 ratio). All patients received preoperative routine ultrasound and contrast-enhanced ultrasound examination, with postoperative pathological confirmation of HCC. Radiomics features were extracted from maximum section of a two-dimensional ultrasound image. The least absolute shrinkage and selection operation logistic regression algorithm with 10-fold cross-validation was used to establish ultrasonic radiomics features. Logistic regression modelling was used to build models based on clinical and ultrasonic features (model 1, clinical-ultrasonic model), radiomics signature (model 2, ultrasonic radiomics model), and the combination (model 3, clinical-ultrasonic-radiomics model). Then, a nomogram model was established to predict the risk of early recurrence, and the application value of nomogram through internal verification was evaluated.

RESULTS

Model 3 showed optimal diagnostic performance in both training set (area under the curve [AUC], 0.907) and validation set (AUC, 0.925), followed by the model 1 in training set (AUC, 0.846) and validation set (AUC, 0.855), both above two models performed better than model 2 in training set (AUC, 0.751) and validation set (AUC, 0.702) (p < 0.05). In the training set and validation set of model 3, the sensitivity were 83.3%, 77.8%, the specificity ware 95.8%, 100.0% and the C-index were 0.791, 0.778.

CONCLUSION

The preoperative clinical-ultrasonic-radiomics model is anticipated to be a reliable tool for predicting the early recurrence of surgical resection of HCC.

HMGB1/TREM1 crosstalk between heat-injured hepatocytes and macrophages promotes HCC progression after RFA

PURPOSE

Tumor recurrence after radiofrequency ablation (RFA) affects the survival rate of patients and limits its clinical application. Tumor recurrence around the ablation area may be related to the thermal injury of hepatocytes (HCs) around the tumor, but the specific mechanism is still unclear.

METHODS

A liver cancer thermal injury mouse model was established via RFA in the C57BL/6 mice. Primary HCs and Kupffer cells (KCs) were isolated and cultured to assess their sensitivity to thermal injury via the MTT assay. Flow cytometry was used to assess macrophage polarization. Furthermore, Western blotting and co-immunoprecipitation (co-IP) were utilized to evaluate the protein expression of intracellular signaling pathway. Finally, Transwell and wound healing assays was conducted to evaluate the invasion potential of liver cancer cells.

RESULTS

Our findings revealed that RFA-induced liver thermal injury promoted the upregulation and secretion of HMGB1 in HCs. HMGB1 had a protective effect on HCs thermal injury, potentially mediated through autophagy regulation. Heat-injured HCs release HMGB1, which activates the TREM1/JAK2/STAT3 signaling pathway in KCs, thus fostering an immunosuppressive tumor microenvironment (TME). Moreover, HMGB1 secretion by heat-injured HCs exacerbates the migration and invasion of HCC cells by influencing macrophage polarization.

CONCLUSION

RFA-induced thermal injury triggers HMGB1 release from HCs, driving macrophage M2 polarization and increasing the invasion ability of liver cancer cells. These findings reveal a potential therapeutic target for combating liver cancer recurrence following thermal ablation.

The role of RNA modifications in hepatocellular carcinoma: functional mechanism and potential applications

Hepatocellular carcinoma (HCC) is a highly aggressive cancer with a poor prognosis. The molecular mechanisms underlying its development remain unclear. Recent studies have highlighted the crucial role of RNA modifications in HCC progression, which indicates their potential as therapeutic targets and biomarkers for managing HCC. In this review, we discuss the functional role and molecular mechanisms of RNA modifications in HCC through a review and summary of relevant literature, to explore the potential therapeutic agents and biomarkers for diagnostic and prognostic of HCC. This review indicates that specific RNA modification pathways, such as N6-methyladenosine, 5-methylcytosine, N7-methylguanosine, and N1-methyladenosine, are erroneously regulated and are involved in the proliferation, autophagy, innate immunity, invasion, metastasis, immune cell infiltration, and drug resistance of HCC. These findings provide a new perspective for understanding the molecular mechanisms of HCC, as well as potential targets for the diagnosis and treatment of HCC by targeting specific RNA-modifying enzymes or recognition proteins. More than ten RNA-modifying regulators showed the potential for use for the diagnosis, prognosis and treatment decision utility biomarkers of HCC. Their application value for HCC biomarkers necessitates extensive multi-center sample validation in the future. A growing number of RNA modifier inhibitors are being developed, but the lack of preclinical experiments and clinical studies targeting RNA modification in HCC poses a significant obstacle, and further research is needed to evaluate their application value in HCC treatment. In conclusion, this review provides an in-depth understanding of the complex interplay between RNA modifications and HCC while emphasizing the promising potential of RNA modifications as therapeutic targets and biomarkers for managing HCC.

Sandwich-type supersensitive electrochemical aptasensor of glypican-3 based on PrGO-Hemin-PdNP and AuNP@PoPD

A new sandwich-type electrochemical biosensing platform was developed by gold @polyphthalenediamine nanohybrids (AuNP@PoPD) as the sensing platform and phosphorus doped reduced graphene oxide-hemin-palladium nanoparticles (PrGO-Hemin-PdNP) as the signal amplifier for phosphatidylinositol proteoglycan 3 (GPC3). AuNP@PoPD, co-electrodeposited into the screen printed electrode with high conductivity and stability, is dedicated to assembling the primary GPC3 aptamer (GPC3Apt). The second GPC3Apt immobilized on the high conductivity and large surface area of PrGO-Hemin-PdNP was utilized as an electrochemical signal reporter by hemin oxidation (PrGO-Hemin-PdNP-GPC3Apt). In the range 0.001-10.0 ng/mL, the hemin oxidation current signal of the electrochemical aptasensor increased log-linearly with the concentration of GPC3, the lowest detection limit was 0.13 pg/mL, and the sensitivity was 2.073 μA/μM/cm2. The aptasensor exhibited good sensing performance in a human serum sample with the relative error of 4.31-8.07%. The sandwich sensor showed good selectivity and stability for detection GPC3 in human serum samples, providing a new efficient and sensitive method for detecting HCC markers.

Novel insights into small open reading frame-encoded micropeptides in hepatocellular carcinoma: A potential breakthrough

Traditionally, non-coding RNAs (ncRNAs) are regarded as a class of RNA transcripts that lack encoding capability; however, advancements in technology have revealed that some ncRNAs contain small open reading frames (sORFs) that are capable of encoding micropeptides of approximately 150 amino acids in length. sORF-encoded micropeptides (SEPs) have emerged as intriguing entities in hepatocellular carcinoma (HCC) research, shedding light on this previously unexplored realm. Recent studies have highlighted the regulatory functions of SEPs in the occurrence and progression of HCC. Some SEPs exhibit inhibitory effects on HCC, but others facilitate its development. This discovery has revolutionized the landscape of HCC research and clinical management. Here, we introduce the concept and characteristics of SEPs, summarize their associations with HCC, and elucidate their carcinogenic mechanisms in HCC metabolism, signaling pathways, cell proliferation, and metastasis. In addition, we propose a step-by-step workflow for the investigation of HCC-associated SEPs. Lastly, we discuss the challenges and prospects of applying SEPs in the diagnosis and treatment of HCC. This review aims to facilitate the discovery, optimization, and clinical application of HCC-related SEPs, inspiring the development of early diagnostic, individualized, and precision therapeutic strategies for HCC.

Risk Models for Predicting the Recurrence and Survival in Patients With Hepatocellular Carcinoma Undergoing Radio-Frequency Ablation

Background

Hepatocellular carcinoma (HCC) patients have a poor prognosis after radio-frequency ablation (RFA), and investigating the risk factors affecting RFA and establishing predictive models are important for improving the prognosis of HCC patients.

Methods

Patients with HCC undergoing RFA in Shenzhen People's Hospital between January 2011 and December 2021 were included in this study. Using the screened independent influences on recurrence and survival, predictive models were constructed and validated, and the predictive models were then used to classify patients into different risk categories and assess the prognosis of different categories.

Results

Cox regression model indicated that cirrhosis (hazard ratio [HR] = 1.65), alpha-fetoprotein (AFP) ⩾400 ng/mL (HR = 2.03), tumor number (multiple) (HR = 2.11), tumor diameter ⩾20 mm (HR = 2.30), and platelets (PLT) ⩾ 244 (109/L) (HR = 2.37) were independent influences for recurrence of patients after RFA. On the contrary, AFP ⩾400 ng/mL (HR = 2.48), tumor number (multiple) (HR = 2.52), tumor diameter ⩾20 mm (HR = 2.25), PLT ⩾244 (109/L) (HR = 2.36), and hemoglobin (HGB) ⩾120 (g/L) (HR = 0.34) were regarded as independent influences for survival. The concordance index (C-index) of the nomograms for predicting disease-free survival (DFS) and overall survival (OS) was 0.727 (95% confidence interval [CI] = 0.770-0.684) and 0.770 (95% CI = 0.821-7.190), respectively. The prognostic performance of the nomograms was significantly better than other staging systems by analysis of the time-dependent C-index and decision curves. Each patient was scored using nomograms and influencing factors, and patients were categorized into low-, intermediate-, and high-risk groups based on their scores. In the Kaplan-Meier survival curve, DFS and OS were significantly better in the low-risk group than in the intermediate- and high-risk groups.

Conclusions

The 2 prediction models created in this work can effectively predict the recurrence and survival rates of HCC patients following RFA.

Exploring the factors affecting the occurrence of postoperative MVI and the prognosis of hepatocellular carcinoma patients treated with hepatectomy: A multicenter retrospective study

OBJECTIVE

To investigate the influencing factors affecting the occurrence of microvascular invasion (MVI) and the prognosis of hepatocellular carcinoma (HCC) patients treated with hepatectomy, and to explore how MVI affects prognosis in subgroups with different prognostic factors.

METHODS

Clinical data of a total of 1633 patients treated surgically for HCC in four treatment centers were included, including 754 patients with MVI. By using the Cox risk regression model and the Mann-Whitney U-test, the common independent influences on prognosis and MVI were made clear. The incidence of MVI in various subgroups was then examined, as well as the relationship between MVI in various subgroups and prognosis.

RESULTS

The Cox risk regression model showed that MVI, Child-Pugh classification, alpha-fetoprotein (AFP), hepatocirrhosis, tumor diameter, lymphocyte-to-monocyte ratio (LMR), and, Barcelona clinic liver cancer (BCLC) grade were independent determinants of overall survival (OS), and MVI, AFP, hepatocirrhosis, tumor diameter, and LMR were influencing determinants for disease-free survival (DFS). The receiver operating characteristic (ROC) curve showed that MVI was most closely associated with patient prognosis compared to other prognostic factors. AFP, hepatocirrhosis, tumor diameter, and LMR were discovered to be common influences on the prognosis of patients with HCC and MVI when combined with the results of the intergroup comparison of MVI. After grouping, it was showed that patients with hepatocirrhosis, positive AFP (AFP ≥ 20 ng/mL), tumor diameter >50 mm, and LMR ≤3.4 had a significantly higher incidence of MVI than patients in other subgroups, and all four subgroups of MVI-positive patients had higher rates of early recurrence and mortality (p < 0.05).

CONCLUSIONS

MVI was found to be substantially linked with four subgroups of HCC patients with hepatocirrhosis, positive AFP, tumor diameter >50 mm, and LMR ≤3.4, and the prognosis of MVI-positive patients in all four subgroups tended to be worse.