Interobserver Variability and Diagnostic Performance of Gadoxetic Acid-enhanced MRI for Predicting Microvascular Invasion in Hepatocellular Carcinoma

MRI Radiomics Combined with Clinicopathological Factors for Predicting 3-Year Overall Survival of Hepatocellular Carcinoma After Hepatectomy

Background

A limited number of studies have examined the use of radiomics to predict 3-year overall survival (OS) after hepatectomy in patients with hepatocellular carcinoma (HCC). This study develops 3-year OS prediction models for HCC patients after liver resection using MRI radiomics and clinicopathological factors.

Materials and Methods

A retrospective analysis of 141 patients who underwent surgical resection of HCC was performed. Patients were randomized into two set: the training set (n=98) and the validation set (n=43) including the survival groups (n=111) and non-survival groups (n=30) based on 3-year survival after hepatectomy. Furthermore, x2 or Fisher's exact test, univariate and multivariate logistic regression analyses were conducted to determine independent clinicopathological risk factors associated with 3-year OS. 1688 quantitative imaging features were extracted from preoperative T2-weighted imaging (T2WI) and contrast-enhanced magnetic resonance imaging (CE-MRI) of arterial phase (AP), portal venous phases (PVP)and delay period (DP). The features were selected using the variance threshold method, the select K best method and the least absolute shrinkage and selection operator (LASSO) algorithm. By using Bernoulli Naive Bayes (BernoulliNB) and Multinomial Naive Bayes (MultinomialNB) classifiers, we constructed models based on the independent clinicopathological factors and Rad-scores. To determine the best model, receiver operating characteristics (ROC) and Delong's test were used. Moreover, calibration curves were used to determine the calibration ability of the model, while decision curve analysis (DCA) was implemented to evaluate its clinical benefit.

Results

The fusion model showed excellent prediction precision with AUC of 0.910 and 0.846 in training and validation set and revealed significant diagnostic accuracy and value in the calibration curve and DCA analysis.

Conclusion

Nomograms based on MRI radiomics and clinicopathological factors have significant predictive value for 3-year OS after hepatectomy and can be used for risk classification.

Local Ablation Therapy for Hepatocellular Carcinoma: Clinical Significance of Tumor Size, Location, and Biology

ABSTRACT

Local ablation therapy, encompassing radiofrequency ablation (RFA), microwave ablation, and cryoablation, has emerged as a crucial strategy for managing small hepatocellular carcinomas (HCCs), complementing liver resection and transplantation. This review delves into the clinical significance of tumor size, location, and biology in guiding treatment decisions for HCCs undergoing local ablation therapy, with a focus on tumors smaller than 3 cm. Tumor size significantly influences treatment outcomes, with larger tumors associated with poorer local tumor control due to challenges in creating sufficient ablative margins and the likelihood of microvascular invasion and peritumoral satellite nodules. Advanced ablation techniques such as centripetal or no-touch RFA using multiple electrodes, cryoablation using multiple cryoprobes, and microwave ablation offer diverse options for HCC treatment. Notably, no-touch RFA demonstrates superior local tumor control compared with conventional RFA by achieving sufficient ablative margins, making it particularly promising for hepatic dome lesions or tumors with aggressive biology. Laparoscopic RFA proves beneficial for treating anterior subphrenic HCCs, whereas artificial pleural effusion-assisted RFA is effective for controlling posterior subphrenic HCCs. However, surgical resection generally offers better survival outcomes for periportal HCCs compared with RFA. Cryoablation exhibits a lower incidence of vascular or biliary complications than RFA for HCCs adjacent to perivascular or periductal regions. Additionally, aggressive tumor biology, such as microvascular invasion, can be predicted using magnetic resonance imaging findings and serum tumor markers. Aggressive HCC subtypes frequently exhibit Liver Imaging Reporting and Data System M features on magnetic resonance imaging, aiding in prognosis. A comprehensive understanding of tumor size, location, and biology is imperative for optimizing the benefits of local ablation therapy in managing HCCs.

Radiomics analysis based on contrast-enhanced MRI for predicting short-term efficacy of drug-eluting beads transarterial chemoembolization in hepatocellular carcinoma

OBJECTIVE

We developed and validated a clinical-radiomics model for preoperative prediction of the short-term efficacy of initial drug-eluting beads transarterial chemoembolization (D-TACE) treatment in patients with hepatocellular carcinoma (HCC).

METHODS

In this retrospective cohort study of 113 patients with intermediate and advanced HCC, 5343 features were extracted based on three sequences of the arterial phase (AP), diffusion-weighted imaging, and T2-weighted images based on contrast-enhanced magnetic resonance imaging, and minimum redundancy maximum correlation and least absolute shrinkage and selection operator (LASSO) regression were applied for feature selection and model construction. Multifactor logistic regression was used to build a clinical-imaging model based on clinical factors and a clinical-radiomics model. The area under the curve (AUC) and calibration curves were used to assess model performance, and the clinical value of the model was analyzed using decision curve analysis. The relationship between the actual and predicted short-term efficacy of the combined model and progression-free survival (PFS) was evaluated using Kaplan-Meier survival curves and log-rank tests.

RESULTS

A total of 34 radiomics features were selected by LASSO, and the clinical-radiomics model had the best predictive performance (AUC = 0.902 and AUC = 0.845 for the training and testing sets, respectively), and the model based on AP had the best predictive performance among the four radiomics models (AUC = 0.89 for the training set and AUC = 0.85 for the testing set); the multifactorial logistic regression results showed that microsphere type (p = 0.042) and AP Rad-score (p = 0.01) were associated with short-term efficacy. In addition, a difference in PFS was observed in patients with HCC with different short-term efficacies predicted by the combined model. Moreover, prognosis was better in the objective versus non-objective response group.

CONCLUSIONS

The combined clinical-radiomics model is an effective predictor of the short-term efficacy of initial D-TACE in patients with HCC, contributing to clinical and economic benefits for patients.

Comparative analysis of the performance of hepatobiliary agents in depicting MRI features of microvascular infiltration in hepatocellular carcinoma

OBJECTIVE

To compare the ability to depict MRI features of hepatobiliary agents in microvascular infiltration (MVI) of hepatocellular carcinoma (HCC) during different stages of dynamic enhancement MRI.

MATERIALS AND METHODS

A retrospective study included 111 HCC lesions scanned with either Gd-EOB-DTPA or Gd-BOPTA. All cases underwent multiphase dynamic contrast-enhanced scanning before surgery, including arterial phase (AP), portal venous phase (PVP), transitional phase (TP), delayed phase (DP), and hepatobiliary phase (HBP). Two abdominal radiologists independently evaluated MRI features of MVI in HCC, such as peritumoral hyperenhancement, incomplete capsule, non-smooth tumor margins, and peritumoral hypointensity. Finally, the results were reviewed by the third senior abdominal radiologist. Chi-square (χ2) Inspection for comparison between groups. P < 0.05 is considered statistically significant. Receiver operating characteristic (ROC) curve was used to evaluate correlation with pathology, and the area under the curve (AUC) and 95% confidence interval (95% CI) were calculated.

RESULTS

Among the four MVI evaluation signs, Gd-BOPTA showed significant differences in displaying two signs in the HBP (P < 0.05:0.000, 0.000), while Gd-EOB-DTPA exhibited significant differences in displaying all four signs (P < 0.05:0.005, 0.006, 0.000, 0.002). The results of the evaluations of the two contrast agents in the DP phase with incomplete capsulation showed the highest correlation with pathology (AUC: 0.843, 0.761). By combining the four MRI features, Gd-BOPTA and Gd-EOB-DTPA have correlated significantly with pathology, and Gd-BOPTA is better (AUC: 0.9312vs0.8712).

CONCLUSION

The four features of hepatobiliary agent dynamic enhancement MRI demonstrate a good correlation with histopathological findings in the evaluation of MVI in HCC, and have certain clinical significance.

Magnetic resonance elastography as a preoperative assessment for predicting intrahepatic recurrence in patients with hepatocellular carcinoma

PURPOSE

Magnetic resonance elastography (MRE) is a noninvasive tool for diagnosing hepatic fibrosis with high accuracy. We investigated the preoperative clinical and imaging predictors of intrahepatic recurrence after curative resection of hepatocellular carcinoma (HCC), and evaluated MRE as a predictor of intrahepatic recurrence.

METHODS

We retrospectively evaluated 80 patients who underwent preoperative contrast-enhanced magnetic resonance imaging (MRI) with two-dimensional MRE and curative resection for treatment-naïve HCC between May 2019 and December 2021. Liver stiffness (LS) was measured on the elastograms, and the optimal cutoff of LS for predicting intrahepatic recurrence was obtained using receiver operating characteristic (ROC) analysis. An LS above this cutoff was defined as MRE-recurrence. Preoperative imaging features of the tumor were assessed on MRI, including features in the Liver Imaging Reporting and Data System and microvascular invasion (MVI). Recurrence-free survival (RFS) rates were estimated using the Kaplan-Meier method, and differences were compared using the log-rank test. Using a Cox proportional hazards model, we conducted a multivariable analysis to investigate the factors affecting recurrence-free survival.

RESULTS

During a median follow-up period of 32 months (range, 4-52 months), thirteen patients (16.3%) developed intrahepatic recurrence. ROC analysis determined an LS cutoff of ≥4.35 kPa to define MRE-recurrence. The 4-year RFS rate was significantly higher in patients without MRE-recurrence than in those with MRE-recurrence (93.4% vs. 48.9%; p = 0.001). In multivariable analysis, MRE-recurrence (Hazard ratio [HR], 5.9; 95% confidence interval [CI], 1.5-23.1) and MVI (HR, 3.4; 95% CI, 1.0-11.3) were independent predictors of intrahepatic recurrence.

CONCLUSIONS

Patients without MRE-recurrence had significantly higher RFS rates than those with MRE-recurrence. MRE-recurrence and MVI were independent predictors of intrahepatic recurrence in patients after curative resection for HCC.

Multiparametric assessment of microvascular invasion in hepatocellular carcinoma using gadoxetic acid-enhanced MRI

PURPOSE

To elucidate how precisely microvascular invasion (MVI) in hepatocellular carcinoma (HCC) can be predicted using multiparametric assessment of gadoxetic acid-enhanced MRI.

METHODS

In this retrospective single-center study, patients who underwent liver resection or transplantation of HCC were evaluated. Data obtained in patients who underwent liver resection were used as the training set. Nine kinds of MR findings for predicting MVI were compared between HCCs with and without MVI by univariate analysis, followed by multiple logistic regression analysis. Using significant findings, a predictive formula for diagnosing MVI was obtained. The diagnostic performance of the formula was investigated in patients who underwent liver resection (validation set 1) and in patients who underwent liver transplantation (validation set 2) using a receiver operating characteristic curve analysis. The area under the curves (AUCs) of these three groups were compared.

RESULTS

A total of 345 patients with 356 HCCs were selected for analysis. Tumor diameter (D) (P = 0.021), tumor washout (TW) (P < 0.01), and peritumoral hypointensity in the hepatobiliary phase (PHH) (P < 0.01) were significantly associated with MVI after multivariate analysis. The AUCs for predicting MVI of the predictive formula were as follows: training set, 0.88 (95% confidence interval (CI) 0.82,0.93); validation set 1, 0.81 (95% CI 0.73,0.87); validation set 2, 0.67 (95% CI 0.51,0.80). The AUCs were not significantly different among three groups (training set vs validation set 1; P = 0.15, training set vs validation set 2; P = 0.09, validation set 1 vs validation set 2; P = 0.29, respectively).

CONCLUSION

Our multiparametric assessment of gadoxetic acid-enhanced MRI performed quite precisely and with good reproducibility for predicting MVI.

Predicting microvascular invasion in hepatocellular carcinoma with a CT- and MRI-based multimodal deep learning model

PURPOSE

To investigate the value of a multimodal deep learning (MDL) model based on computed tomography (CT) and magnetic resonance imaging (MRI) for predicting microvascular invasion (MVI) in hepatocellular carcinoma (HCC).

METHODS

A total of 287 patients with HCC from our institution and 58 patients from another individual institution were included. Among these, 119 patients with only CT data and 116 patients with only MRI data were selected for single-modality deep learning model development, after which select parameters were migrated for MDL model development with transfer learning (TL). In addition, 110 patients with simultaneous CT and MRI data were divided into a training cohort (n = 66) and a validation cohort (n = 44). We input the features extracted from DenseNet121 into an extreme learning machine (ELM) classifier to construct a classification model.

RESULTS

The area under the curve (AUC) of the MDL model was 0.844, which was superior to that of the single-phase CT (AUC = 0.706-0.776, P < 0.05), single-sequence MRI (AUC = 0.706-0.717, P < 0.05), single-modality DL model (AUCall-phase CT = 0.722, AUCall-sequence MRI = 0.731; P < 0.05), clinical (AUC = 0.648, P < 0.05), but not to that of the delay phase (DP) and in-phase (IP) MRI and portal venous phase (PVP) CT models. The MDL model achieved better performance than models described above (P < 0.05). When combined with clinical features, the AUC of the MDL model increased from 0.844 to 0.871. A nomogram, combining deep learning signatures (DLS) and clinical indicators for MDL models, demonstrated a greater overall net gain than the MDL models (P < 0.05).

CONCLUSION

The MDL model is a valuable noninvasive technique for preoperatively predicting MVI in HCC.

MRI-Based Radiomics and Deep Learning in Biological Characteristics and Prognosis of Hepatocellular Carcinoma: Opportunities and Challenges

Hepatocellular carcinoma (HCC) is the fifth most common malignancy and the third leading cause of cancer-related death worldwide. HCC exhibits strong inter-tumor heterogeneity, with different biological characteristics closely associated with prognosis. In addition, patients with HCC often distribute at different stages and require diverse treatment options at each stage. Due to the variability in tumor sensitivity to different therapies, determining the optimal treatment approach can be challenging for clinicians prior to treatment. Artificial intelligence (AI) technology, including radiomics and deep learning approaches, has emerged as a unique opportunity to improve the spectrum of HCC clinical care by predicting biological characteristics and prognosis in the medical imaging field. The radiomics approach utilizes handcrafted features derived from specific mathematical formulas to construct various machine-learning models for medical applications. In terms of the deep learning approach, convolutional neural network models are developed to achieve high classification performance based on automatic feature extraction from images. Magnetic resonance imaging offers the advantage of superior tissue resolution and functional information. This comprehensive evaluation plays a vital role in the accurate assessment and effective treatment planning for HCC patients. Recent studies have applied radiomics and deep learning approaches to develop AI-enabled models to improve accuracy in predicting biological characteristics and prognosis, such as microvascular invasion and tumor recurrence. Although AI-enabled models have demonstrated promising potential in HCC with biological characteristics and prognosis prediction with high performance, one of the biggest challenges, interpretability, has hindered their implementation in clinical practice. In the future, continued research is needed to improve the interpretability of AI-enabled models, including aspects such as domain knowledge, novel algorithms, and multi-dimension data sources. Overcoming these challenges would allow AI-enabled models to significantly impact the care provided to HCC patients, ultimately leading to their deployment for clinical use. LEVEL OF EVIDENCE: 5 TECHNICAL EFFICACY: Stage 2.

Nomogram for Predicting Microvascular Invasion in Hepatocellular Carcinoma Using Gadoxetic Acid-Enhanced MRI and Intravoxel Incoherent Motion Imaging

RATIONALE AND OBJECTIVES

Microvascular invasion (MVI) is an important risk factor in hepatocellular carcinoma (HCC), but it can only be determined through histopathological results. The aim of this study was to develop and validate a nomogram for preoperative prediction MVI in HCC using gadoxetic acid-enhanced magnetic resonance imaging (MRI) and intravoxel incoherent motion imaging (IVIM).

MATERIALS AND METHODS

From July 2017 to September 2022, 148 patients with surgically resected HCC who underwent preoperative gadoxetic acid-enhanced MRI and IVIM were included in this retrospective study. Clinical indicators, imaging features, and diffusion parameters were compared between the MVI-positive and MVI-negative groups using the chi-square test, Mann-Whitney U test, and independent sample t test. Receiver operating characteristic (ROC) curve was used to evaluate the diagnostic performance in predicting MVI. Univariate and multivariate analyses were conducted to identify the significant clinical-radiological variables associated with MVI. Subsequently, a predictive nomogram that integrates clinical-radiological risk factors and diffusion parameters was developed and validated.

RESULTS

Serum alpha-fetoprotein level, tumor size, nonsmooth tumor margin, peritumoral hypo-intensity on hepatobiliary phase (HBP), apparent diffusion coefficient value and D value were statistically significant different between MVI-positive group and MVI-negative group. The results of multivariate analysis identified tumor size (odds ratio [OR], 0.786; 95% confidence interval [CI], 0.675-0.915; P < .01), nonsmooth tumor margin (OR, 2.299; 95% CI, 1.005-5.257; P < .05), peritumoral hypo-intensity on HBP (OR, 2.786; 95% CI, 1.141-6.802; P < .05) and D (OR, 0.293; 95% CI,0.089-0.964; P < .05) was the independent risk factor for the status of MVI. In ROC analysis, the combination of peritumoral hypo-intensity on HBP and D demonstrated the highest area under the curve value (0.902) in prediction MVI status, with sensitivity 92.8% and specificity 87.7%. The nomogram exhibited excellent predictive performance with C-index of 0.936 (95% CI 0.895-0.976) in the patient cohort, and had well-fitted calibration curve.

CONCLUSION

The nomogram incorporating clinical-radiological risk factors and diffusion parameters achieved satisfactory preoperative prediction of the individualized risk of MVI in patients with HCC.

Enhancing preoperative diagnosis of microvascular invasion in hepatocellular carcinoma: domain-adaptation fusion of multi-phase CT images

Objectives

In patients with hepatocellular carcinoma (HCC), accurately predicting the preoperative microvascular invasion (MVI) status is crucial for improving survival rates. This study proposes a multi-modal domain-adaptive fusion model based on deep learning methods to predict the preoperative MVI status in HCC.

Materials and methods

From January 2008 to May 2022, we collected 163 cases of HCC from our institution and 42 cases from another medical facility, with each case including Computed Tomography (CT) images from the pre-contrast phase (PCP), arterial phase (AP), and portal venous phase (PVP). We divided our institution's dataset (n=163) into training (n=119) and test sets (n=44) in an approximate 7:3 ratio. Additionally, we included cases from another institution (n=42) as an external validation set (test1 set). We constructed three single-modality models, a simple concatenated multi-modal model, two current state-of-the-art image fusion model and a multi-modal domain-adaptive fusion model (M-DAFM) based on deep learning methods. We evaluated and analyzed the performance of these constructed models in predicting preoperative MVI using the area under the receiver operating characteristic curve (AUC), decision curve analysis (DCA), and net reclassification improvement (NRI) methods.

Results

In comparison with all models, M-DAFM achieved the highest AUC values across the three datasets (0.8013 for the training set, 0.7839 for the test set, and 0.7454 for the test1 set). Notably, in the test set, M-DAFM's Decision Curve Analysis (DCA) curves consistently demonstrated favorable or optimal net benefits within the 0-0.65 threshold probability range. Additionally, the Net Reclassification Improvement (NRI) values between M-DAFM and the three single-modal models, as well as the simple concatenation model, were all greater than 0 (all p < 0.05). Similarly, the NRI values between M-DAFM and the two current state-of-the-art image fusion models were also greater than 0. These findings collectively indicate that M-DAFM effectively integrates valuable information from multi-phase CT images, thereby enhancing the model's preoperative predictive performance for MVI.

Conclusion

The M-DAFM proposed in this study presents an innovative approach to improve the preoperative predictive performance of MVI.

Extracellular contrast agent-enhanced MRI is as effective as gadoxetate disodium-enhanced MRI for predicting microvascular invasion in HCC

PURPOSE

To compare the performances of gadoxetate disodium-enhanced MRI (EOB-MRI) and extracellular contrast agent-enhanced MRI (ECA-MRI) for predicting microvascular invasion (MVI) in HCC.

MATERIALS AND METHODS

From November 2009 to December 2021, consecutive HCC patients who underwent preoperative contrast-enhanced MRI were retrospectively enrolled into either an ECA-MRI or EOB-MRI cohort. In the ECA-MRI cohort, a preoperative MVI score was constructed in the training dataset using a logistic regression model that evaluated pathological type. In a propensity score-matched testing dataset of the ECA-MRI cohort, the MVI score was validated and compared with a previously proposed EOB-MRI-based MVI score calculated in the EOB-MRI cohort. Time-to-early recurrence survival was evaluated by the Kaplan-Meier method with the log-rank test.

RESULTS

A total of 536 patients were included (478 men; 53 years, interquartile range, 46-62 years), 322 (60.1 %) with pathologically confirmed MVI. Based on the training dataset, independent variables associated with MVI included serum alpha-fetoprotein > 400 ng/ml (odds ratio [OR] = 2.3), infiltrative appearance (OR = 4.9), internal artery (OR = 2.5) and nodule-in-nodule architecture (OR = 2.4), which were incorporated into the ECA-MRI-based MVI score. The testing dataset AUC of the ECA-MRI score was 0.720, which was comparable to that of the EOB-MRI-based MVI score (AUC = 0.721; P =.99). Patients from either the ECA-MRI or the EOB-MRI cohort with model-predicted MVI had significantly shorter time-to-early recurrence than those without MVI (P <.001).

CONCLUSION

Based on the preoperative serum alpha-fetoprotein and three MRI features, ECA-MRI demonstrated comparable performance to EOB-MRI for predicting MVI in HCC.

Risk stratification for overall survival and recurrence-free survival after R0 resection for solitary intrahepatic mass-forming cholangiocarcinoma based on preoperative MRI and clinical features

PURPOSE

This study aimed to establish two nomograms for predicting overall survival (OS) and recurrence-free survival (RFS) in patients with solitary intrahepatic mass-forming cholangiocarcinoma (IMCC) based on preoperative magnetic resonance imaging (MRI) features.

METHODS

This retrospective study included 120 consecutive patients who were diagnosed with solitary IMCC. Preoperative MRI and clinical features were collected. Based on the univariate and multivariate Cox regression analyses, two nomograms were constructed to predict OS and RFS, respectively. The effective performance of the nomograms was evaluated using concordance index (C-index). The prognostic stratification systems for OS and RFS were developed and used to classify patients into high- and low-risk groups.

RESULTS

Suspicious lymph nodes, arterial phase (AP) enhancement patterns, and bile duct dilatation were independent predictors of OS, while suspicious lymph nodes, AP enhancement patterns, and necrosis were independent predictors of RFS. The nomograms achieved the C-index values of 0.705/0.710 for OS and 0.721/0.759 for RFS in the development/validation cohorts, which were significantly higher than those of the T and TNM stages (P < 0.05). Patients were stratified into high- and low-risk groups, the 1-year OS and RFS rates of high-risk patients were poorer than those of patients with low-risk in the development cohort (OS: 93.5% vs 76.3%, P < 0.001; RFS: 74.5% vs 22.4%, P < 0.001). Similar results were observed in the validation cohort.

CONCLUSIONS

Two nomograms were constructed based on preoperative MRI features in patients with solitary IMCC for predicting the OS and RFS and facilitate further prognostic stratification.

A Nomogram of Magnetic Resonance Imaging for Preoperative Assessment of Microvascular Invasion and Prognosis of Hepatocellular Carcinoma

BACKGROUND

Microvascular invasion (MVI) is a predictor of recurrence and overall survival in hepatocellular carcinoma (HCC), the preoperative diagnosis of MVI through noninvasive methods play an important role in clinical treatment.

AIMS

To investigate the effectiveness of radiomics features in evaluating MVI in HCC before surgery.

METHODS

We included 190 patients who had undergone contrast-enhanced MRI and curative resection for HCC between September 2015 and November 2021 from two independent institutions. In the training cohort of 117 patients, MVI-related radiomics models based on multiple sequences and multiple regions from MRI were constructed. An independent cohort of 73 patients was used to validate the proposed models. A final Clinical-Imaging-Radiomics nomogram for preoperatively predicting MVI in HCC patients was generated. Recurrence-free survival was analyzed using the log-rank test.

RESULTS

For tumor-extracted features, the performance of signatures in fat-suppressed T1-weighted images and hepatobiliary phase was superior to that of other sequences in a single-sequence model. The radiomics signatures demonstrated better discriminatory ability than that of the Clinical-Imaging model for MVI. The nomogram incorporating clinical, imaging and radiomics signature showed excellent predictive ability and achieved well-fitted calibration curves, outperforming both the Radiomics and Clinical-Radiomics models in the training and validation cohorts.

CONCLUSIONS

The Clinical-Imaging-Radiomics nomogram model of multiple regions and multiple sequences based on serum alpha-fetoprotein, three MRI characteristics, and 12 radiomics signatures achieved good performance for predicting MVI in HCC patients, which may help clinicians select optimal treatment strategies to improve subsequent clinical outcomes.

Hepatocellular Carcinoma: Optimal Radiological Evaluation before Liver Transplantation

Liver transplantation (LT) is the recommended curative-intent treatment for patients with early or intermediate-stage hepatocellular carcinoma (HCC) who are ineligible for resection. Imaging plays a central role in staging and for selecting the best LT candidates. This review will discuss recent developments in pre-LT imaging assessment, in particular LT eligibility criteria on imaging, the technical requirements and the diagnostic performance of imaging for the pre-LT diagnosis of HCC including the recent Liver Imaging Reporting and Data System (LI-RADS) criteria, the evaluation of the response to locoregional therapy, as well as the non-invasive prediction of HCC aggressiveness and its impact on the outcome of LT. We will also briefly discuss the role of nuclear medicine in the pre-LT evaluation and the emerging role of artificial intelligence models in patients with HCC.

Performances and variability of CT radiomics for the prediction of microvascular invasion and survival in patients with HCC: a matter of chance or standardisation?

OBJECTIVES

To measure the performance and variability of a radiomics-based model for the prediction of microvascular invasion (MVI) and survival in patients with resected hepatocellular carcinoma (HCC), simulating its sequential development and application.

METHODS

This study included 230 patients with 242 surgically resected HCCs who underwent preoperative CT, of which 73/230 (31.7%) were scanned in external centres. The study cohort was split into training set (158 patients, 165 HCCs) and held-out test set (72 patients, 77 HCCs), stratified by random partitioning, which was repeated 100 times, and by a temporal partitioning to simulate the sequential development and clinical use of the radiomics model. A machine learning model for the prediction of MVI was developed with least absolute shrinkage and selection operator (LASSO). The concordance index (C-index) was used to assess the value to predict the recurrence-free (RFS) and overall survivals (OS).

RESULTS

In the 100-repetition random partitioning cohorts, the radiomics model demonstrated a mean AUC of 0.54 (range 0.44-0.68) for the prediction of MVI, mean C-index of 0.59 (range 0.44-0.73) for RFS, and 0.65 (range 0.46-0.86) for OS in the held-out test set. In the temporal partitioning cohort, the radiomics model yielded an AUC of 0.50 for the prediction of MVI, a C-index of 0.61 for RFS, and 0.61 for OS, in the held-out test set.

CONCLUSIONS

The radiomics models had a poor performance for the prediction of MVI with a large variability in the model performance depending on the random partitioning. Radiomics models demonstrated good performance in the prediction of patient outcomes.

CLINICAL RELEVANCE STATEMENT

Patient selection within the training set strongly influenced the performance of the radiomics models for predicting microvascular invasion; therefore, a random approach to partitioning a retrospective cohort into a training set and a held-out set seems inappropriate.

KEY POINTS

• The performance of the radiomics models for the prediction of microvascular invasion and survival widely ranged (AUC range 0.44-0.68) in the randomly partitioned cohorts. • The radiomics model for the prediction of microvascular invasion was unsatisfying when trying to simulate its sequential development and clinical use in a temporal partitioned cohort imaged with a variety of CT scanners. • The performance of the radiomics models for the prediction of survival was good with similar performances in the 100-repetition random partitioning and temporal partitioning cohorts.

MRI Features for Predicting Microvascular Invasion and Postoperative Recurrence in Hepatocellular Carcinoma Without Peritumoral Hypointensity

Purpose

To identify MRI features of hepatocellular carcinoma (HCC) that predict microvascular invasion (MVI) and postoperative intrahepatic recurrence in patients without peritumoral hepatobiliary phase (HBP) hypointensity.

Patients and Methods

One hundred and thirty patients with HCC who underwent preoperative gadoxetate-enhanced MRI and curative hepatic resection were retrospectively reviewed. Two radiologists reviewed all preoperative MR images and assessed the radiological features of HCCs. The ability of peritumoral HBP hypointensity to identify MVI and intrahepatic recurrence was analyzed. We then assessed the MRI features of HCC that predicted the MVI and intrahepatic recurrence-free survival (RFS) in the subgroup without peritumoral HBP hypointensity. Finally, a two-step flowchart was constructed to assist in clinical decision-making.

Results

Peritumoral HBP hypointensity (odds ratio, 3.019; 95% confidence interval: 1.071-8.512; P=0.037) was an independent predictor of MVI. The sensitivity, specificity, positive predictive value, negative predictive value, and AUROC of peritumoral HBP hypointensity in predicting MVI were 23.80%, 91.04%, 71.23%, 55.96%, and 0.574, respectively. Intrahepatic RFS was significantly shorter in patients with peritumoral HBP hypointensity (P<0.001). In patients without peritumoral HBP hypointensity, the only significant difference between MVI-positive and MVI-negative HCCs was the presence of a radiological capsule (P=0.038). Satellite nodule was an independent risk factor for intrahepatic RFS (hazard ratio,3.324; 95% CI: 1.733-6.378; P<0.001). The high-risk HCC detection rate was significantly higher when using the two-step flowchart that incorporated peritumoral HBP hypointensity and satellite nodule than when using peritumoral HBP hypointensity alone (P<0.001).

Conclusion

In patients without peritumoral HBP hypointensity, a radiological capsule is useful for identifying MVI and satellite nodule is an independent risk factor for intrahepatic RFS.

Imaging prognostication and tumor biology in hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is the most common primary liver malignancy, and represents a significant global health burden with rising incidence rates, despite a more thorough understanding of the etiology and biology of HCC, as well as advancements in diagnosis and treatment modalities. According to emerging evidence, imaging features related to tumor aggressiveness can offer relevant prognostic information, hence validation of imaging prognostic features may allow for better noninvasive outcomes prediction and inform the selection of tailored therapies, ultimately improving survival outcomes for patients with HCC.

Clinical Interpretability of Deep Learning for Predicting Microvascular Invasion in Hepatocellular Carcinoma by Using Attention Mechanism

Preoperative prediction of microvascular invasion (MVI) is essential for management decision in hepatocellular carcinoma (HCC). Deep learning-based prediction models of MVI are numerous but lack clinical interpretation due to their "black-box" nature. Consequently, we aimed to use an attention-guided feature fusion network, including intra- and inter-attention modules, to solve this problem. This retrospective study recruited 210 HCC patients who underwent gadoxetate-enhanced MRI examination before surgery. The MRIs on pre-contrast, arterial, portal, and hepatobiliary phases (hepatobiliary phase: HBP) were used to develop single-phase and multi-phase models. Attention weights provided by attention modules were used to obtain visual explanations of predictive decisions. The four-phase fusion model achieved the highest area under the curve (AUC) of 0.92 (95% CI: 0.84-1.00), and the other models proposed AUCs of 0.75-0.91. Attention heatmaps of collaborative-attention layers revealed that tumor margins in all phases and peritumoral areas in the arterial phase and HBP were salient regions for MVI prediction. Heatmaps of weights in fully connected layers showed that the HBP contributed the most to MVI prediction. Our study firstly implemented self-attention and collaborative-attention to reveal the relationship between deep features and MVI, improving the clinical interpretation of prediction models. The clinical interpretability offers radiologists and clinicians more confidence to apply deep learning models in clinical practice, helping HCC patients formulate personalized therapies.

A Reliable and Repeatable Model for Predicting Microvascular Invasion in Patients With Hepatocellular Carcinoma

RATIONALE AND OBJECTIVES

The reproducibility of imaging models for predicting microvascular invasion (MVI) in patients with hepatocellular carcinoma (HCC) remains questionable due to inconsistent interpretation of image signs. Our aim was to screen for high-consensus MRI features to develop a repeatable model for predicting MVI.

MATERIALS AND METHODS

We included 219 patients with HCC who underwent surgical resection, and patients were divided into a training cohort (n = 145) and a validation cohort (n = 74). Morphological characteristics, signal features on hepatobiliary phases, and dynamic enhancement patterns were qualitatively interobserver evaluated. Interobserver agreement was assessed using Cohen's κ for selecting features with high interobserver agreement. Risk factors that were significant in stepwise multivariate analysis and that could be measured with good interobserver agreement were used to construct a predictive model, which was assessed in the validation cohort. The diagnostic performance of the model was evaluated based on area under the receiver operating characteristic curve (AUC).

RESULTS

Multivariate analysis identified nonsmooth tumor margin, absence of radiologic capsule, and intratumoral artery as independent risk factors of MVI. These MRI-based features showed good or nearly perfect interobserver agreement between radiologists (κ > 0.6). The predictive model predicted MVI well in the training (AUC 0.734) and validation cohorts (AUC 0.759) and fitted well to calibration curves.

CONCLUSION

MRI features included nonsmooth tumor margin, absence of radiologic capsule, and intratumoral artery that can be assessed with high interobserver agreement can predict MVI in HCC patients. The predictive model described here may be useful to radiologists, regardless of experience level.

A Nomogram Based on Contrast-Enhanced Ultrasound to Predict the Microvascular Invasion in Hepatocellular Carcinoma

OBJECTIVE

The aim of this study was to establish and validate a contrast-enhanced ultrasound (CEUS) nomogram for pre-operative microvascular invasion (MVI) prediction in hepatocellular carcinoma (HCC), and compare it with the nomogram based on gadopentetate dimeglumine-enhanced magnetic resonance imaging (Gd-MRI).

METHODS

A total of 251 patients with a single HCC were enrolled in this prospective study, including 176 patients in the training cohort and 75 patients in the validation cohort. Contrast-enhanced ultrasound (CEUS) with Sonazoid and Gd-MRI was performed pre-operatively. Post-operative histopathology was the gold standard for MVI. Univariate and multivariate logistic regression was performed to determine independent risk factors for MVI. Nomograms based on CEUS and Gd-MRI were established, and their discrimination, calibration and decision curve analysis were evaluated and compared.

RESULTS

Multivariate logistic regression revealed that arterial circular enhancement, non-enhancing area and thick ring-like enhancement in the post-vascular phase were independent risk factors for MVI. The areas under the receiver operating characteristic curve of the nomogram were 0.841 (0.779-0.892) and 0.914 (0.827-0.966) in the training and validation cohorts, with no significant difference compared with the Gd-MRI nomogram (p = 0.294, 0.321). The C-indexes were 0.821 and 0.870 in the training and validation cohorts. Decision curve analysis revealed that the CEUS nomogram had better clinical applicability than the Gd-MRI nomogram when the threshold probability was between 0.35 and 0.95.

CONCLUSION

The CEUS-based nomogram was available for predicting MVI in HCC, and its predictive performance was not inferior to that of Gd-MRI.

Radiologic features of hepatocellular carcinoma related to prognosis

The cross-sectional imaging findings play a crucial role in the diagnosis of hepatocellular carcinoma (HCC). Recent studies have shown that imaging findings of HCC are not only relevant for the diagnosis of HCC, but also for identifying genetic and pathologic characteristics and determining prognosis. Imaging findings such as rim arterial phase hyperenhancement, arterial phase peritumoral hyperenhancement, hepatobiliary phase peritumoral hypointensity, non-smooth tumor margin, low apparent diffusion coefficient, and the LR-M category of the Liver Imaging-Reporting and Data System have been reported to be associated with poor prognosis. In contrast, imaging findings such as enhancing capsule appearance, hepatobiliary phase hyperintensity, and fat in mass have been reported to be associated with a favorable prognosis. Most of these imaging findings were examined in retrospective, single-center studies that were not adequately validated. However, the imaging findings can be applied for deciding the treatment strategy for HCC, if their significance can be confirmed by a large multicenter study. In this literature, we would like to review imaging findings related to the prognosis of HCC as well as their associated clinicopathological characteristics.

There is no such thing as a validated prediction model

BACKGROUND

Clinical prediction models should be validated before implementation in clinical practice. But is favorable performance at internal validation or one external validation sufficient to claim that a prediction model works well in the intended clinical context?

MAIN BODY

We argue to the contrary because (1) patient populations vary, (2) measurement procedures vary, and (3) populations and measurements change over time. Hence, we have to expect heterogeneity in model performance between locations and settings, and across time. It follows that prediction models are never truly validated. This does not imply that validation is not important. Rather, the current focus on developing new models should shift to a focus on more extensive, well-conducted, and well-reported validation studies of promising models.

CONCLUSION

Principled validation strategies are needed to understand and quantify heterogeneity, monitor performance over time, and update prediction models when appropriate. Such strategies will help to ensure that prediction models stay up-to-date and safe to support clinical decision-making.

VICT2 Trait: Prognostic Alternative to Peritumoral Hepatobiliary Phase Hypointensity in HCC

Background Peritumoral hepatobiliary phase (HBP) hypointensity is an established prognostic imaging feature in hepatocellular carcinoma (HCC), often associated with microvascular invasion (MVI). Similar prognostic features are needed for non-HBP MRI. Purpose To propose a non-hepatobiliary-specific MRI tool with similar prognostic value to peritumoral HBP hypointensity. Materials and Methods From December 2011 to November 2021, consecutive patients with HCC who underwent preoperative contrast-enhanced MRI were retrospectively enrolled and followed up until recurrence. All MRI scans were reviewed by two blinded radiologists with 7 and 10 years of experiences with liver MRI. A scoring system based on non-hepatobiliary-specific features that highly correlated with peritumoral HBP hypointensity was identified in a stratified sampling-derived training set of the gadoxetate disodium (EOB) group by means of multivariable logistic regression, and its values to predict MVI and recurrence-free survival (RFS) were assessed. Results There were 660 patients (551 men; median age, 53 years; IQR, 45-61 years) enrolled. Peritumoral portal venous phase hypoenhancement (odds ratio [OR] = 8.8), incomplete "capsule" (OR = 3.3), corona enhancement (OR, 2.6), and peritumoral mild-moderate T2 hyperintensity (OR, 2.2) (all P < .001) were associated with peritumoral HBP hypointensity and constituted the "VICT2 trait" (test set area under the receiver operating characteristic curve = 0.84; 95% CI: 0.78, 0.90). For the EOB group, both peritumoral HBP hypointensity (OR for MVI = 2.5, P = .02; hazard ratio for RFS = 2.5, P < .001) and the VICT2 trait (OR for MVI = 5.1, P < .001; hazard ratio for RFS = 2.3, P < .001) were associated with MVI and RFS, despite a higher specificity of the VICT2 trait for MVI (89% vs 80%, P = .01). These values of the VICT2 trait were confirmed in the extracellular contrast agent group (OR for MVI = 4.0; hazard ratio for RFS = 1.7; both P < .001). Conclusion Based on four non-hepatobiliary-specific MRI features, the VICT2 trait was comparable to peritumoral hepatobiliary phase hypointensity in predicting microvascular invasion and postoperative recurrence of hepatocellular carcinoma. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Harmath in this issue.

The Organ Procurement and Transplantation Network hepatocellular carcinoma classification: Alignment with Liver Imaging Reporting and Data System, current gaps, and future direction

The Organ Procurement and Transplantation Network (OPTN) updated its allocation policy for liver transplantation to align with the Liver Imaging Reporting and Data System (LI-RADS) for the diagnosis of hepatocellular carcinoma (HCC). LI-RADS computed tomography/magnetic resonance imaging algorithm had achieved congruency with the American Association for the Study of Liver Diseases (AASLD) HCC Practice Guidance in 2018, and therefore, alignment of OPTN, LI-RADS, and AASLD unifies HCC diagnostic approaches. The two changes to the OPTN HCC classification are adoption of LI-RADS terminology or lexicon for HCC major imaging features as well as the modification of OPTN Class-5A through the adoption of LI-RADS-5 criteria. However, despite this significant milestone, the OPTN allocation policy may benefit from further refinements such as adoption of treatment response assessment criteria after locoregional therapy and categorization criteria for lesions with atypical imaging appearances that are not specific for HCC. In this review, we detail the changes to the OPTN HCC classification to achieve alignment with LI-RADS, discuss current limitations of the OPTN classification, and explore future directions.

Risk Group Stratification for Recurrence-Free Survival and Early Tumor Recurrence after Radiofrequency Ablation for Hepatocellular Carcinoma

PURPOSE

Although the prognosis after radiofrequency ablation (RFA) for hepatocellular carcinoma (HCC) may vary according to different risk levels, there is no standardized follow-up protocol according to each patient's risk. This study aimed to stratify patients according to their risk of recurrence-free survival (RFS) and early (≤2 years) tumor recurrence (ETR) after RFA for HCC based on predictive models and nomograms and to compare the survival times of the risk groups derived from the models.

METHODS

Patients who underwent RFA for a single HCC (≤3 cm) between January 2012 and March 2014 (n = 152) were retrospectively reviewed. Patients were classified into low-, intermediate-, and high-risk groups based on the total nomogram points for RFS and ETR, respectively, and compared for each outcome. Restricted mean survival times (RMSTs) in the three risk groups were evaluated for both RFS and ETR to quantitatively evaluate the difference in survival times.

RESULTS

Predictive models for RFS and ETR were constructed with c-indices of 0.704 and 0.730, respectively. The high- and intermediate-risk groups for RFS had an 8.5-fold and 2.9-fold higher risk of events than the low-risk group (both p < 0.001), respectively. The high- and intermediate-risk groups for ETR had a 17.7-fold and 7.0-fold higher risk than the low-risk group (both p < 0.001), respectively. The RMST in the high-risk group was significantly lower than that in the other two groups 9 months after RFA, and that in the intermediate-risk group became lower than that in the low-risk group after 21 months with RFS and 24 months with ETR.

CONCLUSION

Our predictive models were able to stratify patients into three groups according to their risk of RFS and ETR after RFA for HCC. Differences in RMSTs may be used to establish different follow-up protocols for the three risk groups.

A novel multimodal deep learning model for preoperative prediction of microvascular invasion and outcome in hepatocellular carcinoma

BACKGROUND

Accurate preoperative identification of the microvascular invasion (MVI) can relieve the pressure from personalized treatment adaptation and improve the poor prognosis for hepatocellular carcinoma (HCC). This study aimed to develop and validate a novel multimodal deep learning (DL) model for predicting MVI based on multi-parameter magnetic resonance imaging (MRI) and contrast-enhanced computed tomography (CT).

METHODS

A total of 397 HCC patients underwent both CT and MRI examinations before surgery. We established the radiological models (R^CT, R^MRI) by support vector machine (SVM), DL models (DL^CT_ALL, DL^MRI_ALL, DL^CT + MRI) by ResNet18. The comprehensive model (C^ALL) involving multi-modality DL features and clinical and radiological features was constructed using SVM. Model performance was quantified by the area under the receiver operating characteristic curve (AUC) and compared by net reclassification index (NRI) and integrated discrimination improvement (IDI).

RESULTS

The DL^CT + MRI model exhibited superior predicted efficiency over single-modality models, especially over the DL^CT_ALL model (AUC: 0.819 vs. 0.742, NRI > 0, IDI > 0). The DL^MRI_ALL model improved the performance over the R^MRI model (AUC: 0.794 vs. 0.766, NRI > 0, IDI < 0), but no such difference was found between the DL^CT_ALL model and R^CT model (AUC: 0.742 vs. 0.710, NRI < 0, IDI < 0). Furthermore, both the DL^CT + MRI and C^ALL models revealed the prognostic power in recurrence-free survival stratification (P < 0.001).

CONCLUSION

The proposed DL^CT + MRI model showed robust capability in predicting MVI and outcomes for HCC. Besides, the identification ability of the multi-modality DL model was better than any single modality, especially for CT.

Prediction for Aggressiveness and Postoperative Recurrence of Hepatocellular Carcinoma Using Gadoxetic Acid-Enhanced Magnetic Resonance Imaging

RATIONALE AND OBJECTIVES

To investigate the predictive value of gadoxetic acid-enhanced magnetic resonance imaging (MRI) features on the pathologic grade, microvascular invasion (MVI), and cytokeratin-19 (CK19) expression in hepatocellular carcinomas (HCC), and to evaluate their association with postoperative recurrence of HCC.

MATERIALS AND METHODS

This retrospective study included 147 patients with surgically confirmed HCCs who underwent gadoxetic-enhanced MRI. The lesions were evaluated quantitatively in terms of the relative enhancement ratio (RER), and qualitatively based on imaging features and clinical parameters. Logistic regression analyses were performed to investigate the value of these parameters in predicting the pathologic grade, MVI, and CK19 in HCC. Predictive factors for postoperative recurrence were determined using a Cox proportional hazards model.

RESULTS

Peritumoral enhancement (odds ratio [OR], 3.396; p = 0.025) was an independent predictor of high pathologic grades. Serum protein induced by vitamin K absence or antagonist (PIVKA) level > 40 mAU/mL (OR, 3.763; p = 0.018) and peritumoral hypointensity (OR, 4.343; p = 0.003) were independent predictors of MVI. Predictors of CK19 included serum alpha-fetoprotein (AFP) level > 400 ng/mL (OR, 4.576; p = 0.005), rim enhancement (OR, 5.493; p = 0.024), and lower RER (OR, 0.013; p = 0.011). Peritumoral hypointensity (hazard ratio [HR], 1.957; p = 0.027) and poor pathologic grades (HR, 2.339; p = 0.043) were independent predictors of recurrence.

CONCLUSION

We demonstrated the value of preoperative gadoxetic-enhanced MRI in predicting aggressive pathological features of HCC. Poor pathologic grades and peritumoral hypointensity may independently predict the recurrence of HCC.

The value of quantitative MR elastography-based stiffness for assessing the microvascular invasion grade in hepatocellular carcinoma

OBJECTIVES

To evaluate the potential diagnostic value of MR elastography (MRE)-based stiffness to noninvasively predict the microvascular invasion (MVI) grade in hepatocellular carcinoma (HCC).

METHODS

One hundred eighty-five patients with histopathology-proven HCC who underwent MRI and MRE examinations before hepatectomy were retrospectively enrolled. According to the three-tiered MVI grading system, the MVI was divided into negative-MVI (n = 89) and positive-MVI (n = 96) groups, and the latter group was categorized into mild-MVI (n = 49) and severe-MVI (n = 47) subgroups. Logistic regression and area under the receiver operating characteristic curve (AUC) analyses were used to determine the predictors associated with MVI grade and analyze their performances, respectively.

RESULTS

Among the 185 patients, tumor size ≥ 50 mm (p = 0.031), tumor stiffness (TS)/liver stiffness (LS) > 1.47 (p = 0.001), TS > 4.33 kPa (p < 0.001), and nonsmooth tumor margin (p = 0.006) were significant independent predictors for positive-MVI. Further analyzing the subgroups, tumor size ≥ 50 mm (p < 0.001), TS > 5.35 kPa (p = 0.001), and AFP level > 400 ng/mL (p = 0.044) were independently associated with severe-MVI. The models incorporating MRE and clinical-radiological features together performed better for evaluating positive-MVI (AUC: 0.846) and severe-MVI (AUC: 0.802) than the models using clinical-radiological predictors alone (AUC: positive-/severe-MVI, 0.737/0.743). Analysis of recurrence-free survival and overall survival showed the predicted positive-MVI/severe-MVI groups based on combined models had significantly poorer prognoses than predicted negative-MVI/mild-MVI groups, respectively (all p < 0.05).

CONCLUSIONS

MRE-based stiffness was an independent predictor for both the positive-MVI and severe-MVI. The combination of MRE and clinical-radiological models might be a useful tool for evaluating HCC patients' prognoses underwent hepatectomy by preoperatively predicting the MVI grade.

KEY POINTS

• The severe-microvascular invasion (MVI) grade had the highest tumor stiffness (TS), followed by mild-MVI and non-MVI, and there were significances among the three different MVI grades. • MR elastography (MRE)-based stiffness value was an independent predictor of positive-MVI and severe-MVI in hepatocellular carcinoma (HCC) preoperatively. • When combined with clinical-radiological models, MRE could significantly improve the predictive performance for MVI grade. Patients with predicted positive-MVI/severe-MVI based on the combined models had worse recurrence-free survival and overall survival than those with negative-MVI/mild-MVI, respectively.

Efficacy of texture analysis of pre-operative magnetic resonance imaging in predicting microvascular invasion in hepatocellular carcinoma

BACKGROUND

Presence of microvascular invasion (MVI) indicates poorer prognosis post-curative resection of hepatocellular carcinoma (HCC), with an increased chance of tumour recurrence. By present standards, MVI can only be diagnosed post-operatively on histopathology. Texture analysis potentially allows identification of patients who are considered ‘high risk’ through analysis of pre-operative magnetic resonance imaging (MRI) studies. This will allow for better patient selection, improved individualised therapy (such as extended surgical margins or adjuvant therapy) and pre-operative prognostication.

AIM

This study aims to evaluate the accuracy of texture analysis on pre-operative MRI in predicting MVI in HCC.

METHODS

Retrospective review of patients with new cases of HCC who underwent hepatectomy between 2007 and 2015 was performed. Exclusion criteria: No pre-operative MRI, significant movement artefacts, loss-to-follow-up, ruptured HCCs, previous hepatectomy and adjuvant therapy. Fifty patients were divided into MVI (n = 15) and non-MVI (n = 35) groups based on tumour histology. Selected images of the tumour on post-contrast-enhanced T1-weighted MRI were analysed. Both qualitative (performed by radiologists) and quantitative data (performed by software) were obtained. Radiomics texture parameters were extracted based on the largest cross-sectional area of each tumor and analysed using MaZda software. Five separate methods were performed. Methods 1, 2 and 3 exclusively made use of features derived from arterial, portovenous and equilibrium phases respectively. Methods 4 and 5 made use of the comparatively significant features to attain optimal performance.

RESULTS

Method 5 achieved the highest accuracy of 87.8% with sensitivity of 73% and specificity of 94%.

CONCLUSION

Texture analysis of tumours on pre-operative MRI can predict presence of MVI in HCC with accuracies of up to 87.8% and can potentially impact clinical management.

Can a proposed double branch multimodality-contribution-aware TripNet improve the prediction performance of the microvascular invasion of hepatocellular carcinoma based on small samples?

OBJECTIVES

To evaluate the potential improvement of prediction performance of a proposed double branch multimodality-contribution-aware TripNet (MCAT) in microvascular invasion (MVI) of hepatocellular carcinoma (HCC) based on a small sample.

METHODS

In this retrospective study, 121 HCCs from 103 consecutive patients were included, with 44 MVI positive and 77 MVI negative, respectively. A MCAT model aiming to improve the accuracy of deep neural network and alleviate the negative effect of small sample size was proposed and the improvement of MCAT model was verified among comparisons between MCAT and other used deep neural networks including 2DCNN (two-dimentional convolutional neural network), ResNet (residual neural network) and SENet (squeeze-and-excitation network), respectively.

RESULTS

Through validation, the AUC value of MCAT is significantly higher than 2DCNN based on CT, MRI, and both imaging (P < 0.001 for all). The AUC value of model with single branch pretraining based on small samples is significantly higher than model with end-to-end training in CT branch and double branch (0.62 vs 0.69, p=0.016, 0.65 vs 0.83, p=0.010, respectively). The AUC value of the double branch MCAT based on both CT and MRI imaging (0.83) was significantly higher than that of the CT branch MCAT (0.69) and MRI branch MCAT (0.73) (P < 0.001, P = 0.03, respectively), which was also significantly higher than common-used ReNet (0.67) and SENet (0.70) model (P < 0.001, P = 0.005, respectively).

CONCLUSION

A proposed Double branch MCAT model based on a small sample can improve the effectiveness in comparison to other deep neural networks or single branch MCAT model, providing a potential solution for scenarios such as small-sample deep learning and fusion of multiple imaging modalities.

Prognostic implications of CT/MRI LI-RADS in hepatocellular carcinoma: State of the art and future directions

Hepatocellular carcinoma (HCC) is the fourth most lethal malignancy with an increasing incidence worldwide. Management of HCC has followed several clinical staging systems that rely on tumour morphologic characteristics and clinical variables. However, these algorithms are unlikely to profile the full landscape of tumour aggressiveness and allow accurate prognosis stratification. Noninvasive imaging biomarkers on computed tomography (CT) or magnetic resonance imaging (MRI) exhibit a promising prospect to refine the prognostication of HCC. The Liver Imaging Reporting and Data System (LI-RADS) is a comprehensive system for standardizing the terminology, techniques, interpretation, reporting and data collection of liver imaging. At present, it has been widely accepted as an effective diagnostic system for HCC in at-risk patients. Emerging data have provided new insights into the potential of CT/MRI LI-RADS in HCC prognostication, which may help refine the prognostic paradigm of HCC that promises to direct individualized management and improve patient outcomes. Therefore, this review aims to summarize several prognostic imaging features at CT/MRI for patients with HCC; the available evidence regarding the use of LI-RDAS for evaluation of tumour biology and clinical outcomes, pitfalls of current literature, and future directions for LI-RADS in the management of HCC.

Diagnostic Accuracy of the Apparent Diffusion Coefficient for Microvascular Invasion in Hepatocellular Carcinoma: A Meta-analysis

BACKGROUND AND AIMS

Microvascular invasion (MVI) is a major risk factor for the early recurrence of hepatocellular carcinoma (HCC) and it seriously worsens the prognosis. Accurate preoperative evaluation of the presence of MVI could greatly benefit the treatment management and prognosis prediction of HCC patients. The study aim was to evaluate the diagnostic performance of the apparent diffusion coefficient (ADC), a quantitative parameter for the preoperative diagnosis MVI in HCC patients.

METHODS

Original articles about diffusion-weighted imaging (DWI) and/or intravoxel incoherent motion (IVIM) conducted on a 3.0 or 1.5 Tesla magnetic resonance imaging (MRI) system indexed through January 17, 2021were collected from MEDLINE/PubMed, Web of Science, EMBASE, and the Cochrane Library. Methodological quality was evaluated using Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2). The pooled sensitivity, specificity, and summary area under the receiver operating characteristic curve (AUROC) were calculated, and meta-regression analysis was performed using a bivariate random effects model through a meta-analysis.

RESULTS

Nine original articles with a total of 988 HCCs were included. Most studies had low bias risk and minimal applicability concerns. The pooled sensitivity, specificity and AUROC of the ADC value were 73%, 70%, and 0.78, respectively. The time interval between the index test and the reference standard was identified as a possible source of heterogeneity by subgroup meta-regression analysis.

CONCLUSIONS

Meta-analysis showed that the ADC value had moderate accuracy for predicting MVI in HCC. The time interval accounted for the heterogeneity.

Radiological Features of Microvascular Invasion of Hepatocellular Carcinoma in Patients with Non-Alcoholic Fatty Liver Disease

Background: The aim of the present study was to evaluate the presence and the prognostic value of the radiological signs of microvascular invasion (MVI) of hepatocellular carcinoma (HCC) in patients with non-alcoholic fatty liver disease (NAFLD). Methods: Between January 2015 and December 2017, all patients (91 patients) with de novo HCC or HCC recurrence occurring at least 2 years after the last treatment in NAFLD (36 patients) or with hepatitis C virus (HCV) liver disease (55 patients) were included. Each HCC was treated with liver resection and transplantation to obtain the anatomopathological confirmation of MVI. All patients had at least one available computed tomography (CT) scan or magnetic resonance imaging (MRI) performed no more than one month prior to the treatment. The clinical data of each patient, tumor burden (diameter, margins, two-trait predictor of venous invasion (TTPVI), and peritumoral enhancement), the recurrence rate (RR) after a 1-year follow-up, and the time to recurrence (TTR) were collected. Results: The NAFLD–HCC nodules were larger as compared to HCV–HCC (51 mm vs. 36 mm, p = 0.004) and showed a higher prevalence of TTPVI (38.9 vs. 20.0%, p = 0.058). At multivariate analysis, nodule diameter >50 mm was found to be the only independent prognostic factor of TTPVI (hazard ratio: 21.3, 95% confidence interval: 4.2–107.7, p < 0.001), and the presence of TTPVI was confirmed to be the only independent prognostic factors of recurrence (hazard ratio: 2.349, 95% confidence interval: 1.369–4.032, p = 0.002). No correlations were found between TTR and irregular tumor margins or peritumoral enhancement. Conclusion: The NAFLD–HCC patients had larger tumors at diagnosis and showed a more frequent presence of radiological signs of MVI as compared to the HCV–HCC patients. The MVI was related to a more rapid recurrence after curative treatments, demonstrating the prognostic value of this radiological diagnosis.

Multi-frequency magnetic resonance elastography of the pancreas: measurement reproducibility and variance among healthy volunteers

Background

Patients with chronic pancreatitis often have irreversible pancreatic insufficiency before a clinical diagnosis. Pancreatic cancer is a fatal malignant tumor in the advanced stages. Patients having high risk of pancreatic diseases must be screened early to obtain better outcomes using new imaging modalities. Therefore, this study aimed to investigate the reproducibility of tomoelastography measurements for assessing pancreatic stiffness and fluidity and the variance among healthy volunteers.

Methods

Forty-seven healthy volunteers were prospectively enrolled and underwent two tomoelastography examinations at a mean interval of 7 days. Two radiologists blindly and independently measured the pancreatic stiffness and fluidity at the first examination to determine the reproducibility between readers. One radiologist measured the adjacent pancreatic slice at the first examination to determine the reproducibility among slices and measured the pancreas at the second examination to determine short-term repeatability. The stiffness and fluidity of the pancreatic head, body, and tail were compared to determine anatomical differences. The pancreatic stiffness and fluidity were compared based on sex, age, and body mass index (BMI).

Results

Bland–Altman analyses (all P > 0.05) and intraclass correlation coefficients (all >0.9) indicated near perfect reproducibility among readers, slices, and examinations at short intervals. Neither stiffness (P = 0.477) nor fluidity (P = 0.368) differed among the pancreatic anatomical regions. The mean pancreatic stiffness was 1.45 ± 0.09 m/s; the mean pancreatic fluidity was 0.83 ± 0.06 rad. Stiffness and fluidity did not differ by sex, age, or BMI.

Conclusion

Tomoelastography is a promising and reproducible tool for assessing pancreatic stiffness and fluidity in healthy volunteers.

Evaluation of perfusion CT and dual-energy CT for predicting microvascular invasion of hepatocellular carcinoma

PURPOSE

Evaluation of perfusion CT and dual-energy CT (DECT) quantitative parameters for predicting microvascular invasion (MVI) of hepatocellular carcinoma (HCC) prior to surgery.

METHODS

This prospective single-center study included fifty-six patients (44 men; median age 67; range 31-84) who provided written informed consent. Inclusion criteria were (1) treatment-naïve patients with a diagnosis of HCC, (2) an indication for hepatic resection, and (3) available arterial DECT phase and perfusion CT (GE revolution HD-GSI). Iodine concentrations (IC), arterial density (AD), and 9 quantitative perfusion parameters for HCC were correlated to pathological results. Radiological parameters based principal component analysis (PCA), corroborated by unsupervised heatmap classification, was meant to deliver a model for predicting MVI in HCC. Survival analysis was performed using univariable log-rank test and multivariable Cox model, both censored at time of relapse.

RESULTS

58 HCC lesions were analyzed (median size 42.3 mm; range of 20-140). PCA showed that the radiological model was predictive of tumor grade (p = 0.01), intratumoral MVI (p = 0.004), peritumoral MVI (p = 0.04), MTM (macrotrabecular-massive) subtype (p = 0.02), and capsular invasion (p = 0.02) in HCC. Heatmap classification of HCC showed tumor heterogeneity, stratified into three main clusters according to the risk of relapse. Survival analysis confirmed that permeability surface-area product (PS) was the only significant independent parameter, among all quantitative tumoral CT parameters, for predicting a risk of relapse (Cox p value = 0.004).

CONCLUSION

A perfusion CT and DECT-based quantitative imaging profile can provide a diagnosis of histological MVI in HCC. PS is an independent parameter for relapse.

CLINICAL TRIALS

ClinicalTrials.gov: NCT03754192.

Prediction of microvascular invasion in hepatocellular carcinoma with expert-inspiration and skeleton sharing deep learning

BACKGROUND AND AIMS

Radiological prediction of microvascular invasion (MVI) of hepatocellular carcinoma (HCC) is essential but few models were clinically implemented because of limited interpretability and generalizability.

METHODS

Based on 2096 patients in three independent HCC cohorts, we established and validated an MVI predicting model. First, we used data from the primary cohort to train a 3D-ResNet network for MVI prediction and then optimised the model with "expert-inspired training" for model construction. Second, we implemented the model to the other two cohorts using three implementation strategies, the original model implementation, data sharing model implementation and skeleton sharing model implementation, the latter two of which used part of the cohorts' data for fine-tuning. The areas under the receiver operating characteristic curve (AUCs) were calculated to compare the performances of different models.

RESULTS

For the MVI predicting model, the AUC of the expert-inspired model was 0.83 (95% CI: 0.77-0.88) compared to 0.54 (95% CI: 0.46-0.62) of model before expert-inspiring. Taking this model as an original model, AUC on the second cohort was 0.76 (95% CI: 0.67-0.84). The AUC was improved to 0.83 (95% CI: 0.77-0.90) with the data-sharing model, and further improved to 0.85 (95% CI: 0.79-0.92) with the skeleton sharing model. The trend that the skeleton sharing model had an advantage in performance was similar in the third cohort.

CONCLUSIONS

We established an expert-inspired model with better predictive performance and interpretability than the traditional constructed model. Skeleton sharing process is superior to data sharing and direct model implementation in model implementation.

Gadoxetate-Enhanced MRI as a Diagnostic Tool in the Management of Hepatocellular Carcinoma: Report from a 2020 Asia-Pacific Multidisciplinary Expert Meeting

Gadoxetate magnetic resonance imaging (MRI) is widely used in clinical practice for liver imaging. For optimal use, we must understand both its advantages and limitations. This article is the outcome of an online advisory board meeting and subsequent discussions by a multidisciplinary group of experts on liver diseases across the Asia-Pacific region, first held on September 28, 2020. Here, we review the technical considerations for the use of gadoxetate, its current role in the management of patients with hepatocellular carcinoma (HCC), and its relevance in consensus guidelines for HCC imaging diagnosis. In the latter part of this review, we examine recent evidence evaluating the impact of gadoxetate on clinical outcomes on a continuum from diagnosis to treatment decision-making and follow-up. In conclusion, we outline the potential future roles of gadoxetate MRI based on an evolving understanding of the clinical utility of this contrast agent in the management of patients at risk of, or with, HCC.

Predicting microvascular invasion in hepatocellular carcinoma: A dual-institution study on gadoxetate disodium-enhanced MRI

BACKGROUND & AIMS

Microvascular invasion (MVI) is an important risk factor in hepatocellular carcinoma (HCC), but its diagnosis mandates postoperative histopathologic analysis. We aimed to develop and externally validate a predictive scoring system for MVI.

METHODS

From July 2015 to November 2020, consecutive patients underwent surgery for HCC with preoperative gadoxetate disodium (EOB)-enhanced MRI was retrospectively enrolled. All MR images were reviewed independently by two radiologists who were blinded to the outcomes. In the training centre, a radio-clinical MVI score was developed via logistic regression analysis against pathology. In the testing centre, areas under the receiver operating curve (AUCs) of the MVI score and other previous MVI schemes were compared. Overall survival (OS) and recurrence-free survival (RFS) were analysed by the Kaplan-Meier method with the log-rank test.

RESULTS

A total of 417 patients were included, 195 (47%) with pathologically-confirmed MVI. The MVI score included: non-smooth tumour margin (odds ratio [OR] = 4.4), marked diffusion restriction (OR = 3.0), internal artery (OR = 3.0), hepatobiliary phase peritumoral hypointensity (OR = 2.5), tumour multifocality (OR = 1.6), and serum alpha-fetoprotein >400 ng/mL (OR = 2.5). AUCs for the MVI score were 0.879 (training) and 0.800 (testing), significantly higher than those for other MVI schemes (testing AUCs: 0.648-0.684). Patients with model-predicted MVI had significantly shorter OS (median 61.0 months vs not reached, P < .001) and RFS (median 13.0 months vs. 42.0 months, P < .001) than those without.

CONCLUSIONS

A preoperative MVI score integrating five EOB-MRI features and serum alpha-fetoprotein level could accurately predict MVI and postoperative survival in HCC. Therefore, this score may aid in individualized treatment decision making.

LI-RADS Version 2018 Targetoid Appearances on Gadoxetic Acid-Enhanced MRI: Interobserver Agreement and Diagnostic Performance for the Differentiation of HCC and Non-HCC Malignancy

Background: In LI-RADS version 2018, observations showing at least one of five targetoid appearances on different sequences or postcontrast phases are assigned LR-M, indicating likely non-hepatocellular carcinoma (HCC) malignancy. Objective: To evaluate interobserver agreement of the LI-RADS targetoid appearances among a large number of radiologists of varying experiences, and the targetoid appearances' diagnostic performance for differentiating HCC from non-HCC malignancy. Methods: This retrospective study included 100 patients (76 men, 24 women; mean age, 58±9 years) at high-risk for HCC who underwent gadoxetic acid-enhanced MRI within 30 days before hepatic tumor resection [25 randomly selected patients with non-HCC malignancy (13 intrahepatic cholangiocarcinoma, 12 combined HCC-cholangiocarcinoma); 75 matched patients with HCC]. Eight radiologists [four more-experienced (8-15 years); four less-experienced (1-5 years)] from seven different institutions independently assessed observations for the five targetoid appearances and LI-RADS categorization. Interobserver agreement and diagnostic performance for non-HCC malignancy were evaluated. Results: Interobserver agreement was poor for peripheral washout (κ=0.20); moderate for targetoid transitional-phase or hepatobiliary-phase appearance (κ=0.33), delayed central enhancement (κ=0.37), and targetoid restriction (κ=0.43); and substantial for rim arterial-phase hyperenhancement (κ=0.61). Agreement was fair for at least one targetoid appearance (κ=0.36) and moderate for at least two, three, or four targetoid appearances (κ=0.43-0.51). Agreement for individual targetoid appearances was not significantly different between more-experienced and less-experienced readers other than for targetoid restriction (κ=0.63 vs 0.43; p=.001). Agreement for at least one targetoid appearance was fair among more-experienced (κ=0.29) and less-experienced (κ=0.37) reviewers. Agreement for at least two, three, or four targetoid appearances was moderate-to-substantial among more-experienced reviewers (κ=0.45-0.63) and moderate among less-experienced reviewers (κ=0.42-0.56). Existing LR-M criteria of at least one targetoid appearance had median accuracy for non-HCC malignancy of 62%, sensitivity of 84%, and specificity of 54%. For all reviewers, accuracy was highest when requiring at least three (median accuracy 79%, sensitivity 68%, specificity 82%) or four (median accuracy 80%, sensitivity 54%, specificity 88%) targetoid appearances. Conclusion: Targetoid appearances and LR-M categorization exhitibted considerable interobserver variation for both more- and less-experienced reviewers. Clinical Impact: Requirement for multiple targetoid appearances for LR-M categorization improved interobserver agreement and diagnostic accuracy for non-HCC malignancy.

The role of gadoxetic acid-enhanced MRI features for predicting microvascular invasion in patients with hepatocellular carcinoma

OBJECTIVE

To evaluate the predictive value of gadoxetic acid-enhanced MRI features (focused on Liver Imaging Reporting and Data System (LI-RADS) v2018 features and non-LI-RADS imaging features) for microvascular invasion (MVI) of hepatocellular carcinoma (HCC).

METHODS

From October 2018 to December 2020, 134 patients who underwent gadoxetic acid-enhanced MRI with a pathological diagnosis of HCC after hepatectomy were enrolled in this retrospective study. Two radiologists assessed the pre-hepatectomy LI-RADS v2018 imaging features and non-LI-RADS features to identify independent predictors of MVI of HCC with a logistic regression model.

RESULTS

Four MRI features were found to be independent predictors of MVI: corona enhancement [odds ratio (OR) 5.787; 95% confidence interval (CI) 1.180, 28.369; p = 0.030], mosaic architecture (OR 7.097; 95% CI 1.299, 38.783; p = 0.024), nonsmooth tumor margin (OR 13.131; 95% CI 3.950, 43.649; p < 0.001), and peritumoral hypointensity on hepatobiliary phase (HBP) (OR 33.123; 95% CI 2.897, 378.688; p = 0.005). When one of four imaging features was present, the sensitivity was 93.2% (41/44), and the specificity was 71.1% (64/90).

CONCLUSION

The four imaging features including corona enhancement, mosaic architecture, nonsmooth tumor margin, and peritumoral hypointensity on HBP can be used as preoperative imaging biomarkers for predicting MVI in patients at high risk for HCC. When one of the four imaging features is present, MVI can be predicted with a sensitivity > 90%.

MR Elastography-Based Shear Strain Mapping for Assessment of Microvascular Invasion in Hepatocellular Carcinoma

OBJECTIVES

To evaluate the potential of MR elastography (MRE)-based shear strain mapping to noninvasively predict the presence of microvascular invasion (MVI) in hepatocellular carcinoma (HCC).

METHODS

Fifty-nine histopathology-proven HCC patients with conventional 60-Hz MRE examinations (+/-MVI, n = 34/25) were enrolled retrospectively between December 2016 and October 2019, with one subgroup comprising 29/59 patients (+/-MVI, n = 16/13) who also underwent 40- and 30-Hz MRE examinations. Octahedral shear strain (OSS) maps were calculated, and the percentage of peritumoral interface length with low shear strain (i.e., a low-shear-strain length, pLSL, %) was recorded. For OSS-pLSL, differences between the MVI (+) and MVI (-) groups and diagnostic performance at different MRE frequencies were analyzed using the Mann-Whitney test and area under the receiver operating characteristic curve (AUC), respectively.

RESULTS

The peritumor OSS-pLSL was significantly higher in the MVI (+) group than in the MVI (-) group at the three frequencies (all p < 0.01). The AUC of peritumor OSS-pLSL for predicting MVI was good/excellent in all frequency groups (60-Hz: 0.73 (n = 59)/0.80 (n = 29); 40-Hz: 0.84; 30-Hz: 0.90). On further analysis of the 29 cases with all frequencies, the AUCs were not significantly different. As the frequency decreased from 60-Hz, the specificity of OSS increased at 40-Hz (53.8-61.5%) and further increased at 30-Hz (53.8-76.9%), and the sensitivity remained high at lower frequencies (100.0-93.8%) (all p > 0.05).

CONCLUSIONS

MRE-based shear strain mapping is a promising technique for noninvasively predicting the presence of MVI in patients with HCC, and the most recommended frequency for OSS is 30-Hz.

KEY POINTS

• MR elastography (MRE)-based shear strain mapping has the potential to predict the presence of microvascular invasion (MVI) in hepatocellular carcinoma preoperatively. • The low interface shear strain identified at tumor-liver boundaries was highly correlated with the presence of MVI.

Noninvasive Visualization of Sub-5 mm Orthotopic Hepatic Tumors by a Nanoprobe-Mediated Positive and Reverse Contrast-Balanced Imaging Strategy

Delineation of small malignant lesions and their vasculature enables early and accurate diagnosis of hepatocellular carcinoma (HCC). However, it remains challenging to identify these features simultaneously by noninvasive imaging technology. Reverse contrast imaging emerges as a powerful means to detect early-stage HCC by taking inspiration from the intrinsic liver phagocytosis toward exogenous agents to generate negative tumor-to-normal tissue signals. However, this mechanism conflicts with the signal-enhancing requirements for vasculature visualization. Here, we solve this conundrum by designing a positive and reverse contrast-balanced imaging strategy based on a multifunctional PEG-Ta₂O₅@CuS nanoprobe that combines advanced gemstone spectral computer tomography (GSCT) with photoacoustic (PA) imaging. The nanoprobe exhibits preferential accumulation in Kupffer cells and hepatocytes over tumor cells, and its spectral properties are well matched with GSCT, leading to the enhancement of reverse contrast signals that enable clear delineation of 2-4 mm orthotopic HCC lesions. Meanwhile, its strong PA imaging capability at the second near-infrared (NIR-II) window makes vascular evaluation accessible by monitoring the positive signal enhancement derived from the limited tumor accumulation of the nanoprobe. In addition, the nanoprobe enables NIR-II photohyperthermia for timely tumor ablation. Overall, this proposed strategy shows potential in early detection and theranostics of HCC for improved clinical outcomes.

A Non-Invasive Nomogram for Preoperative Prediction of Microvascular Invasion Risk in Hepatocellular Carcinoma

Background

Microvascular invasion (MVI) is a significant predictive factor for early recurrence, metastasis, and poor prognosis of hepatocellular carcinoma. The aim of the present study is to identify preoperative factors for predicting MVI, in addition to develop and validate non-invasive nomogram for predicting MVI.

Methods

A total of 381 patients with resected HCC were enrolled and divided into a training cohort (n = 267) and a validation cohort (n = 114). Serum VEGF-A level was examined by enzyme-linked immunosorbent assay (ELISA). Risk factors for MVI were assessed based on univariate and multivariate analyses in the training cohort. A nomogram incorporating independent risk predictors was established and validated.

Result

The serum VEGF-A levels in the MVI positive group (n = 198) and MVI negative group (n = 183) were 215.25 ± 105.68 pg/ml and 86.52 ± 62.45 pg/ml, respectively (P <0.05). Serum VEGF-A concentration ≥138.30 pg/ml was an independent risk factor of MVI (OR: 33.088; 95%CI: 12.871–85.057; P <0.001). Higher serum concentrations of AFP and VEGF-A, lower lymphocyte count, peritumoral enhancement, irregular tumor shape, and intratumoral artery were identified as significant predictors for MVI. The nomogram indicated excellent predictive performance with an AUROC of 0.948 (95% CI: 0.923–0.973) and 0.881 (95% CI: 0.820–0.942) in the training and validation cohorts, respectively. The nomogram showed a good model fit and calibration.

Conclusions

Higher serum concentrations of AFP and VEGF-A, lower lymphocyte count, peritumoral enhancement, irregular tumor shape, and intratumoral artery are promising markers for MVI prediction in HCC. A reliable non-invasive nomogram which incorporated blood biomarkers and imaging risk factors was established and validated. The nomogram achieved desirable effectiveness in preoperatively predicting MVI in HCC patients.