Abstract Hepatocellular carcinoma (HCC) frequently coexists with portal hypertension, significantly increasing the risk of hepatic decompensation (HD) and variceal bleeding during systemic therapy. We developed a machine learning based hepatic safety score (MHSS) using data from 2026 patients with unresectable HCC to predict clinically significant portal hypertension (CSPH) and prognosis. A random forest model was trained in a derivation cohort (n = 1262) and validated in an independent cohort (n = 764). The MHSS demonstrated robust performance (AUROC 0.840) in CSPH and predicting HD. Stratification revealed that high MHSS patients faced significantly elevated risks of HD (HR 3.25), variceal bleeding (VB, HR 4.90), and mortality (HR 2.21). Crucially, while atezolizumab-bevacizumab offered a survival advantage in low MHSS patients, it was associated with high bleeding risk and no survival benefit in the high MHSS group compared to other regimens. A simulation of MHSS guided treatment selection demonstrated a 24% reduction in HD, a 40% reduction in VB, and a 26% reduction in mortality. In conclusion, the MHSS effectively predicts CSPH, decompensation, and survival in patients with HCC prior to systemic therapy. By enabling individualized risk stratification, the MHSS may guide personalized treatment selection between bevacizumab-containing and alternative regimens, ultimately improving patient outcomes. Clinical trial number: not applicable.

Bile duct tumor thrombus (BDTT) is a rare manifestation of hepatocellular carcinoma (HCC) that causes obstructive jaundice and is associated with a poor prognosis, although a treatment algorithm has yet to be established. We review the treatment landscape for HCC with BDTT, including surgical, transarterial, systemic, and radiotherapy options, and report, to our knowledge, the first case series of magnetic resonance-guided radiotherapy (MRgRT) for this condition. Four patients with unresectable HCC and BDTT were treated on a 0.35-T MR-linac with real-time cine-MRI gating (50 Gy in 5 fractions, n = 3; 40 Gy in 5 fractions, n = 1), with tumor response assessed by modified RECIST. All patients completed treatment without interruption. The best response was complete response in two patients and partial response in two, and obstructive jaundice resolved in both affected patients. The maximum radiation-attributed toxicity was a transient grade 3 bilirubin elevation that resolved without biliary intervention, and no treatment-related deaths occurred. Overall survival ranged from 5.5 to 29 months, with the two Child-Pugh class A patients without portal vein tumor thrombosis surviving 22 and 29 months. MRgRT for HCC with BDTT appears feasible with acceptable short-term safety and merits prospective evaluation.

The therapeutic paradigm for advanced hepatocellular carcinoma has shifted with the global adoption of immune checkpoint inhibitor-based regimens, leading to the frequent use of multi-tyrosine kinase inhibitors in later treatment lines. This study aimed to evaluate the real-world efficacy and safety of regorafenib as a later-line salvage therapy compared with its conventional second-line application within contemporary clinical practice. This study included 255 patients with advanced hepatocellular carcinoma treated with regorafenib between 2017 and 2026. Clinical outcomes, including objective response rate, disease control rate, overall survival (OS), progression-free survival (PFS), and time to progression, were compared between patients receiving regorafenib as a second-line versus later-line therapy. Of the 255 enrolled patients, 209 received regorafenib as a second-line therapy and 46 as a later-line therapy. The mean maintained dose was significantly lower in the later-line group than in the second-line group. The overall objective response rate and disease control rate were 11.0

Background/Aims:Hepatic decompensation (HD) following systemic treatment, including atezolizumab + bevacizumab (Atezo/Bev) and tyrosine kinase inhibitors (TKIs), is a critical prognostic event in advanced hepatocellular carcinoma (HCC). This study aimed to evaluate the predictive utility of liver stiffness measurement (LSM) by vibration-controlled transient elastography (VCTE) for HD incidence post-treatment. Methods:This multicenter study included 396 HCC patients who received systemic therapy (Atezo/Bev or TKIs) and underwent VCTE prior to treatment at seven university-affiliated hospitals. Clinical outcomes including HD independent of tumor progression, variceal bleeding (VB), overall survival (OS), and progression-free survival (PFS) were assessed. A 25 kPa LSM threshold, based on Baveno VII criteria, stratified patients into high and low LSM groups. Results:Of the 396 patients, 176 received Atezo/Bev, while 45 and 175 received lenvatinib and sorafenib, respectively. Treatment distribution was similar between high and low LSM groups (p = 0.546). High LSM was associated with increased HD risk (HR = 3.00, p < 0.001), VB risk (HR = 2.34, p = 0.048), and a trend toward worse OS (HR = 1.27, p = 0.065). In the low LSM group, Atezo/Bev outperformed TKIs in OS and PFS (p < 0.05) without increasing HD risk. In contrast, in the high LSM group, Atezo/Bev and TKIs showed no OS or PFS differences, but Atezo/Bev significantly increased HD and VB risk (p < 0.05). A risk score based on four variables (Child-Pugh score 5, LSM ≥25 kPa, multiple tumors, and high-grade portal vein tumor thrombosis) showed good predictive accuracy for HD at 12 months (AUC = 0.832) and effectively identified patients at high risk for HD, VB, and poor survival (p < 0.005). Conclusion:LSM by VCTE predicts HD following systemic treatment in advanced HCC. In patients with high LSM, Atezo/Bev increases HD risk, warranting careful treatment selection.

Despite advances in hepatitis C virus (HCV) treatments, diagnostic gaps and linkage-to-care hinder elimination efforts. We evaluated an electronic medical record (EMR)-based automated alert system to improve HCV care at tertiary referral centers. We have screened 1,303,578 patients who were tested for anti-HCV, and analyzed 8291 patients positive for anti-HCV antibody (Ab) across four tertiary hospitals between July 2009 and December 2023. The EMR alert system, implemented in June 2021, identified patients with positive anti-HCV antibody results who had not undergone HCV ribonucleic acid (RNA) testing. We compared HCV RNA testing rates, linkage-to-care and time intervals between testing before and after system implementation. The HCV RNA test prescription rate increased from 60.5% in the pre-alert period to 66.9% in the post-alert period. In non-hepatology departments, the prescription rate increased from 47.1% to 60.5%. The interval between HCV Ab positivity to HCV RNA test prescription in patients without initial HCV RNA order decreased significantly from 1208.9 to 244.5 days. Among 732 previously HCV RNA untested patients who were followed up after alert implementation, 43.4% received HCV RNA testing, leading to 62 new HCV diagnoses. The referral rate to hepatology departments remained stable (83.3% pre-alert vs. 80.9% post-alert). In conclusion, implementation of an EMR-based alert system effectively improved HCV diagnostic rates and reduced testing delays, particularly in non-hepatology departments. This system represents a promising strategy for hospital-based HCV micro-elimination, but additional interventions might be needed to achieve optimal testing rates and linkage-to-care.

Background & Aims Hepatic decompensation is a key determinant of prognosis in chronic liver disease (CLD). Shear wave elastography (SWE) is widely used to assess fibrosis, but its prognostic value for decompensation remains uncertain. We evaluated SWE alone and combined with ALBI for risk prediction. Methods In this retrospective two-center cohort study, 2,759 patients with CLD who underwent SWE between 2019 and 2025 were analyzed. The endpoint was time to first decompensation event. Predictive performance of SWE, alone and with ALBI, was evaluated in training (n=1,839) and test (n=920) sets. Results During a median follow-up of 22.5 months, a total of 181 decompensation events and 53 deaths occurred. SWE alone showed good performance for predicting decompensation (C-index 0.765) and OS. A 10-kPa threshold effectively stratified risk, with high-risk patients having higher hazards of decompensation and death (P<0.001). SWE and ALBI showed no statistical interaction, supporting their potential for complementary synergistic effects when combined. Integrating SWE with ALBI grade created a simple three-tier classification that further improved stratification, with high-risk patients (SWE >10 kPa and ALBI grade 2/3) having a 16.6-fold risk of decompensation and a 21.4-fold risk of death compared with low-risk patients (P<0.001). A Fine–Gray model in the training set generated a weighted linear formula (SWE + 18×ALBI), termed the SALBI score, achieving the highest accuracy (C-index: training 0.804, test 0.820), outperforming SWE, ALBI, and MELD. The SALBI score stratified high- versus low-risk groups with 11.5-fold and 14.3-fold risks in train and test sets (P<0.001). Conclusions SWE is a robust prognostic marker for hepatic decompensation. A linear score combining SWE with ALBI further improves prediction and provides a practical tool for stratification.

Background/Aims: cccDNA maintains HBV persistence through its transcriptional activity. While cccDNA and pgRNA are essential for replication, serum markers reflecting active cccDNA remain poorly defined across chronic hepatitis B (CHB) phases. We aimed to identify serum markers correlated with intrahepatic cccDNA and its activity. Methods: Samples from 406 CHB patients were analyzed by disease phase. Serum (hepatitis B surface antigen [HBsAg], hepatitis B core-related antigen [HBcrAg], and HBV RNA) and intrahepatic (covalently closed circular DNA [cccDNA] and pregenomic RNA [pgRNA]) markers were quantified using commercial assays. Results: All serum and intrahepatic markers were inversely associated with age and were higher in HBeAg-positive than in HBeAg-negative patients. Correlations among serum and intrahepatic markers were stronger in HBeAg-positive patients. Serum markers correlated more strongly with pgRNA levels than with cccDNA quantity, with HBcrAg and HBV RNA showing stronger associations with intrahepatic markers than HBsAg. Intrahepatic cccDNA transcriptional activity (pgRNA/cccDNA) varied across CHB phases, being highest in the immune-tolerant phase, lowest in the inactive-carrier (IC) phase, moderate in the immune-active (IA) phase, and re-elevated in HBeAg-negative hepatitis (ENH). Newer markers better reflected cccDNA transcriptional activity than conventional marker across the IA and ENH phases. Notably, no serum marker reliably reflected pgRNA or transcriptional activity in the IC phase. Conclusions: Serum HBcrAg and HBV RNA represent reliable surrogates for transcriptionally active cccDNA in a phase-dependent manner. The absence of reliable markers in the IC phase highlights its virologic heterogeneity. These findings support their phase-specific use for personalized monitoring and treatment.

The long-term prognosis of immune tolerant (IT) phase patients remains unclear. This study aimed to identify true IT-phase patients with favourable outcomes and distinguish grey-zone IT patients. We retrospectively analysed 1064 chronic hepatitis B (CHB) patients (516 clinically inclusive IT, 548 antiviral-treated immune active [AVT-IA]). Clinically inclusive IT-phase was defined as HBeAg-positive, no cirrhosis, HBV DNA ≥ 10^6 IU/mL, and ALT < 80 U/L. Favourable prognosis was defined as the absence of HCC development and IA progression during follow-up. IT-phase patients had significantly lower 10-year HCC rates compared to AVT-IA patients (1.7% vs. 2.7%). Among the clinically inclusive IT group, favourable prognosis was associated with younger age (< 35 years), female gender, stringent low ALT (< 35 U/L for males, < 25 U/L for females), high HBV DNA (> 10^8 IU/mL), and no family history of HCC. Patients meeting these criteria showed no HCC cases and had a lower likelihood of progression to IA (~50%) within 10 years. The strict criteria demonstrated 90.3% specificity for identifying true IT-phase patients within the clinically inclusive IT group. Family history of HCC was an independent risk factor for HCC development (HR 6.059, p = 0.019), while stringent low ALT, younger age, and female gender were linked to lower IA progression risk (HR 0.32, p ≤ 0.001). This study highlights the importance of applying strict criteria in distinguishing true IT-phase patients from grey-zone patients. True IT patients are at minimal risk of HCC or progression to IA, while identifying grey-zone patients is crucial, as they may require early antiviral therapy.

BACKGROUND AND AIMS:Atezolizumab plus bevacizumab (AB) has become the standard first-line treatment for advanced HCC. However, identifying reliable prognostic biomarkers remains a critical challenge. We aimed to develop a comprehensive scoring system to predict overall survival (OS) in advanced HCC patients receiving first-line AB. APPROACH AND RESULTS:We included patients with advanced HCC receiving first-line AB from multiple centers in Korea, forming a derivation cohort ( n =456) and a validation cohort ( n =205). Multivariable analysis identified 5 independent prognostic factors: C-reactive protein ≥1.0 mg/dL (HR 2.07; p <0.001), albumin <3.5 g/dL (HR 1.60; p =0.002), protein induced by vitamin K absence or antagonist-II ≥1500 mAU/mL (HR 1.60; p =0.002), total bilirubin ≥1.0 mg/dL (HR 1.50; p =0.006), and macrovascular invasion (HR 1.49; p =0.009). We developed the CRAPT-M model, named after these factors' initial letters. Patients were categorized into low (≤4), intermediate (5-12), and high (≥13) risk groups by CRAPT-M score. Median OS differed significantly: 22.4 (95% CI, 18.6-25.0), 12.9 (95% CI, 8.7-14.8), and 6.7 (95% CI, 5.1-7.7) months for low-risk, intermediate-risk, and high-risk groups, respectively ( p <0.001). Time-dependent area under the receiver operating characteristic for CRAPT-M demonstrated consistently higher predictive accuracy than the CRAFITY model, with values of 0.785, 0.737, and 0.742 at 12, 24, and 36 months, respectively. The model demonstrated robust predictive performance in the external validation cohort, with excellent calibration and consistent discrimination across sensitivity analyses. CONCLUSIONS:The CRAPT-M model demonstrated robust OS prediction, offering a valuable tool for prognosis estimation and clinical decision-making in advanced HCC patients receiving AB.

Background/Aims: Besifovir (BSV) showed comparable antiviral activity and superior safety profiles to tenofovir disoproxil fumarate (TDF) in treatment-na & iuml;ve chronic hepatitis B (CHB). However, no data are available regarding the antiviral efficacy and safety of BSV in patients with CHB who switched from long-term TDF to BSV. This study aimed to evaluate the outcome of a 48-week BSV therapy in patients with CHB who switched from long-term TDF treatment. Methods: In this non-inferiority trial, 153 CHB patients treated with TDF for >= 48 weeks who had hepatitis B virus (HBV) DNA <20 IU/mL were randomized to receive either BSV 150 mg or TDF 300 mg for 48 weeks. Results: The per-protocol analysis included 130 patients (BSV group, 64; TDF group, 66). The median duration of TDF use before enrollment was 4.14 years. After 48 weeks, 100.0% and 98.5% patients in the BSV and TDF groups, respectively, met the primary endpoint (HBV DNA <20 IU/mL), demonstrating the non-inferior antiviral efficacy of BSV to TDF (95% confidence interval-0.01 to 0.04; P>0.999), with a predefined margin of-0.18. The mean percentage changes in estimated glomerular filtration rates were slightly better in the BSV group (1.67 +/- 11.73%) than in the TDF group (-1.24 +/- 11.02%). The BSV group showed a significant improvement in bone turnover biomarkers compared to the TDF group; accordingly, hip and spine bone mineral density increased in the BSV group. Conclusions: In patients with CHB receiving long-term TDF, switching to BSV may improve renal and bone safety with non-inferior antiviral efficacy compared to that of maintaining TDF.

Supplementary Figures Figure S1. Risk analysis of overall survival based on ALBI score. (A) Restricted cubic spline curve showing the association between ALBI score and hazard ratio for overall survival in patients with Child-Pugh score 7. (B) Kaplan–Meier survival analysis comparing overall survival between patients with Child-Pugh score 7 stratified by ALBI score (cut-off: −1.9). Figure S2 Restricted mean survival time analysis in patients with Child-Pugh score 6 and 7. (A) Comparison of RMST between patients with Child-Pugh score 6 and favorable Child-Pugh score 7. (B) Comparison of RMST between patients with Child-Pugh score 6 and unfavorable Child-Pugh score 7. (C) Comparison of RMST between favorable and unfavorable Child-Pugh score 7 groups. Supplementary Tables Table S1. Factors associated with survival outcomes Table S2. Documented adverse events in patients with Child-Pugh score 6 and 7.

BACKGROUND/AIMS:Acute decompensation (AD) is defined as the development of complications related to portal hypertension and liver dysfunction that affect the progression of chronic liver disease (CLD) or liver cirrhosis (LC). Variations exist in patient demographics and prognostic outcomes of AD based on the aetiology of CLD, encompassing LC. However, limited research has been conducted to analyse these discrepancies across aetiologies. METHODS:The prospective Korean Acute-on-Chronic Liver Failure (KACLiF) cohort consisted of 1,501 patients who were hospitalized with AD of CLD from July 2015 to August 2018. In this study, we assess the clinical attributes and prognostic implications of AD with CLD/LC stratified by the aetiology. RESULTS:Among 1,501 patients, the mean age was 54.7 years old and 1,118 patients (74.5%) were men. The common events of AD were GI bleeding (35.3%) and jaundice (35.0%). There was a median follow-up of 8.0 months (1.0-16.0 months). The most common aetiology of CLD was alcohol (n = 1021), followed by viral hepatitis (n = 206), viral hepatitis with alcohol-related (n = 129), cryptogenic (n = 108) and autoimmune (n = 37). Viral hepatitis with alcohol-related CLD showed a poor liver function profile and a high frequency of acute-on-chronic liver failure (ACLF) [22.1% vs. 19.6% (alcohol CLD), 8.1% (viral CLD), 5.6% (autoimmune related CLD and 16.0% (cryptogenic CLD)] with worse adverse outcomes (mortality or liver transplantation) than other aetiologies. The difference in aetiology was a significant factor for 28-day adverse outcomes in multivariate analysis even in a high MELD score (≥15), which indicated poor baseline liver function and prognosis (p < 0.001). CONCLUSION:The aetiology of CLD constitutes a pivotal determinant influencing both short- and long-term adverse outcomes of AD in CLD, even among individuals presenting with elevated MELD scores. Notably, patients afflicted with viral hepatitis should exercise caution even in the consumption of modest quantities of alcohol that induced the exacerbations in the adverse outcomes associated with AD.

Background:Several serum cytokines have been proposed as biomarkers for predicting the outcomes of patients with hepatocellular carcinoma (HCC) receiving tyrosine kinase inhibitors. However, their role in atezolizumab plus bevacizumab (AB) treatment needs to be more elucidated. Methods:We examined various serum cytokines, including interferon-γ (IFN-γ), interleukin-10 (IL-10), IL-12, IL-17, IL-2, IL-6, and tumor necrosis factor, using a Luminex cytokine multiplex assay before AB treatment in prospectively enrolled 116 AB-treatment patients for the derivation cohort and 54 patients for the external validation cohort. We collected baseline characteristics, including neutrophil-lymphocyte ratio (NLR) and C-reactive protein (CRP) levels, and prospectively observed clinical outcomes. Results:Among various peripheral blood inflammatory markers, high NLR, CRP, IL-2, and IL-12 levels were significantly associated with poor progression-free survival (PFS) and overall survival (OS) in patients with AB-treated HCC. Through sensitivity analysis, we defined the high peripheral blood inflammatory score (PBIS) group, which included two or more of the following elevated factors: NLR, CRP, IL-2, and IL-12. The high PBIS group had elevated serum inflammatory cytokines and a higher tumor burden than the low PBIS group. A high PBIS score was an independent risk factor associated with poor OS, PFS, and objective response rate (ORR) in multivariate analyses, which was also confirmed in the validation cohort and propensity score-matched cohort. However, it was not a significant factor for OS, PFS, or ORR in lenvatinib-treated patients. Conclusion:These results suggest that a peripheral blood marker-based scoring system can significantly predict clinical outcomes in patients with AB-treated HCC. This non-invasive biomarker is expected to be a potential predictive and prognostic factor for AB treatment.