Soon Sun Kim

The role of liver biopsy in hepatocellular carcinoma (HCC) is being re-evaluated in the era of precision oncology and biomarker-driven systemic therapy. Although histopathological examination remains the diagnostic gold standard, HCC is unique among solid tumors in that accurate non-invasive imaging criteria allow diagnosis without routine biopsy in selected clinical settings. However, increasing biological heterogeneity, expanding therapeutic options, and the growing need for molecular stratification have renewed interest in tissue-based evaluation. Recent international guidelines have gradually broadened the indications for liver biopsy, particularly in selected diagnostic scenarios, prior to systemic therapy, and for molecular profiling. Beyond diagnostic confirmation, liver biopsy provides important information for histological subtyping, prognostic stratification, and biomarker discovery. Advances in immunohistochemistry, next-generation sequencing, and artificial intelligence-assisted pathology have further enhanced the clinical value of tumor tissue. In parallel, contemporary cohort studies indicate that liver biopsy is associated with low complication rates when modern techniques and standardized peri-procedural management are applied. Despite these advances, the integration of liver biopsy into routine clinical practice remains limited. To contextualize this gap, the Korean Liver Cancer Association conducted a nationwide survey of HCC specialists. The survey showed that liver biopsy is infrequently performed in current practice due to perceived procedural risks and limited immediate impact on therapeutic decision-making. Taken together, accumulating evidence supports a reappraisal of the clinical role of liver biopsy in HCC. Optimizing procedural safety, standardizing indications, and integrating tissue-based molecular insights into therapeutic decision-making may help bridge the gap between emerging evidence and real-world practice.

Background: Women with hepatocellular carcinoma (HCC) present with more favorable tumors and better unadjusted survival than men. Whether this advantage persists within stage, and whether aggregate treatment rates obscure sex differences after case-mix adjustment, is uncertain. Methods: In this nationwide cohort study of the Korean Primary Liver Cancer Registry (2008–2020), 12,118 of 19,668 adults with newly diagnosed HCC had complete prespecified data. Survival was assessed with stage-stratified Kaplan-Meier analyses and sequentially adjusted Cox models. Treatment non-receipt was assessed with logistic regression adjusted for age, stage, Child-Pugh score, performance status, and diagnosis period, with marginal standardization for adjusted probabilities. Findings: 2,608 (21·5%) patients were women, who were older but more often had early-stage disease (67·1% vs 55·4%; p65 years had higher odds (1·270, 1·064–1·517). Interpretation: The apparent female survival advantage in HCC was largely explained by stage distribution. Similar aggregate treatment rates masked modest sex differences after case-mix adjustment and an age-dependent reversal in curative treatment receipt. Sex differences in HCC care should be assessed within case mix, age, and stage strata.

Hyo Jung Cho, Seong Hee Kang, Hye Yeon Chon, SungA Bae, Eunju Kim, Seong Kyun Na, Su Jin Kim, Dooyeon Kim, Hye Ri Ahn, Huigyeong Kim, Soon Sun Kim, Jae Youn Cheong, Soon Koo Baik, Young Kul Jung, Hyung Joon Yim. Gut and Liver 2026;20:870-9. https://doi.org/10.5009/gnl250585

Background and Objective: Accurate prediction of survival outcomes in patients with liver cirrhosis is challenging due to the complex nature of disease progression. Radiomic analysis of routinely acquired computed tomography can expose structural disease burden that laboratory-based scores do not represent, so multimodal fusion of imaging with clinical data holds promise. However, dominant modalities tend to suppress weaker ones during joint training, a phenomenon known as modality laziness, which limits the contribution of imaging. This study proposes an interpretable multimodal survival prediction framework that fuses clinical features with three-dimensional computed tomography (3D CT) imaging for time-to-event prediction in cirrhotic patients.,Methods: The framework addresses modality imbalance through auxiliary unimodal regularization with alternating adaptation. For explainability, we developed Survival Activation Maps, a gradient-free attribution method quantifying the contribution of each spatial feature to predicted risk, since gradient-based attribution is unreliable in multimodal survival models. Performance was evaluated in 3,354 patients across hepatocellular carcinoma (HCC)-free, liver transplantation (LT)-free, and HCC/LT-free survival using the concordance index, the integrated Brier score, and the time-dependent area under the curve, with Kaplan–Meier analysis, log-rank tests, and Cox proportional hazards regression on the held-out test set.,Results: The proposed framework significantly outperformed clinical-only and conventional multimodal baselines across all three outcomes (all p<0.01), with concordance indices of 0.723, 0.805, and 0.755 versus 0.706, 0.789, and 0.737 for conventional multimodal fusion, confirming modality laziness. Survival Activation Maps localized consistently to hepatic and peri-hepatic regions, and albumin and age emerged as the dominant clinical predictors, indicating that 3D CT imaging contributed substantively to survival prediction.,Conclusions: Conventional multimodal integration improved only marginally over clinical-only models, whereas explicitly counteracting modality laziness through alternating unimodal adaptation produced consistent gains and enabled the imaging encoder to contribute meaningfully. By capturing structural features that biochemical markers do not represent, the framework also outperformed four established clinical risk scores. Balanced and explainable multimodal fusion of 3D CT imaging with clinical data therefore enables more individualized risk stratification in cirrhosis than laboratory-based scores, and may identify high-risk patients whose risk is underestimated by preserved liver function tests, supporting intensified surveillance and earlier transplant evaluation.

Purpose: This review summarizes current evidence on the role of the gut microbiome in the pathogenesis, diagnosis, and treatment of metabolic dysfunction-associated steatotic liver disease (MASLD), with particular emphasis on the gut–liver axis and microbiome-based therapeutic approaches.Current concepts: MASLD is among the most prevalent chronic liver diseases worldwide and is strongly associated with cardiometabolic risk factors. Evidence from clinical and preclinical studies indicates that gut microbial dysbiosis contributes to the development and progression of MASLD through increased intestinal permeability, endotoxin translocation, and altered production of microbiota-derived products and metabolites, including lipopolysaccharide, bile acids, short-chain fatty acids, trimethylamine N-oxide, and tryptophan metabolites. Interactions between host genetic susceptibility and the gut microbiome may further influence disease pathogenesis. Compared with healthy individuals, patients with MASLD show reduced microbial diversity and characteristic changes in bacterial, viral, and fungal communities that correlate with disease severity. Shotgun metagenomics and multi-kingdom analyses have identified microbiome-based signatures with potential value as noninvasive diagnostic biomarkers. Therapeutic strategies targeting the gut microbiome, including dietary interventions, probiotics, synbiotics, and fecal microbiota transplantation, have also shown promise, although evidence of histological improvement remains limited.Discussion and conclusion: The gut microbiome plays an important role in MASLD and may inform biomarker development and therapeutic targeting. Future research should prioritize standardized, multicenter, prospective studies and precision-medicine approaches that integrate microbial, genetic, and metabolic profiles to support the development of clinically applicable microbiome-based strategies.

Background: Women with hepatocellular carcinoma (HCC) present with more favorable tumors and better survival than men, but whether this reflects a within-stage benefit or a favorable stage distribution is unclear. Methods: We analyzed the 12,118 newly diagnosed HCC patients using Korean Primary Liver Cancer Registry (2008–2020). Stage-stratified Kaplan–Meier analysis and sequentially adjusted Cox and logistic regression were used, adjusting for age, modified UICC (mUICC) stage, Child–Pugh score, performance status and diagnosis period. Findings: Of 12,118 patients, 2,608 (21.5%) were women. Women were older but had smaller tumors, less macrovascular invasion and more early-stage disease (mUICC I–II, 67.1% vs 55.4%; P < .001), yet crude treatment non-receipt was similar (13.6% vs 13.0%; P = .43). Survival curves overlapped at every stage, and the unadjusted female advantage (HR 0.829, 95% CI 0.782–0.878) attenuated to 0.855 (0.807–0.907) after adjustment. After full adjustment, women had higher odds of treatment non-receipt (OR 1.180, 95% CI 1.015–1.373; P = .032; adjusted absolute difference +1.4 percentage points), and this finding was more apperent in women aged >65 and non-alcohol, non-viral etiologies (ORs for non-receipt 1.49 [1.11-1.99] and 1.27 [1.06-1.52]). Among early-stage patients, women aged ≤65 years were more likely and women aged >65 years less likely than men to receive curative treatment (ORs for non-receipt 0.84 [0.72–0.98] and 1.27 [1.06–1.52]). Interpretation: Women with HCC presented with more favorable tumor characteristics, and adjustment for stage and liver function markedly attenuated their observed survival advantage. Despite similar crude treatment rates, women had higher adjusted odds of treatment nonreceipt. These findings indicate that aggregate treatment rates may obscure sex-related differences in HCC care after accounting for clinical case mix.

While long noncoding RNAs (lncRNAs) play critical roles in hepatocellular carcinoma (HCC) pathogenesis, the regulatory mechanisms governing liver‐specific lncRNAs remain poorly defined. Here, we systematically analyzed staging‐specific transcriptomic datasets to identify functionally relevant lncRNAs involved in HCC progression. We identified a liver‐specific lncRNA, termed LUNAR (Liver‐specific Upregulated Non‐coding RNA as Down‐Regulator in HCC), reflecting its high expression in normal liver tissue and its functional role in restraining tumor progression in HCC. Analyses across multiple independent patient cohorts revealed that LUNAR expression correlated with clinical outcomes in HCC. Functional assays demonstrated that restoration of LUNAR expression suppressed cell migration, invasion, and in vivo metastasis, without affecting cell proliferation. Mechanistically, LUNAR overexpression was associated with attenuation of NOTCH signaling, while its downregulation in tumors was linked to epigenetic silencing via promoter hypermethylation. Collectively, these findings identify LUNAR as a liver‐enriched, epigenetically regulated lncRNA with metastasis‐restraining functions. This therefore supports the potential of LUNAR as a tissue‐associated biomarker and therapeutic target in HCC.

The role of direct-acting antivirals (DAAs) in patients with hepatocellular carcinoma (HCC) remains uncertain due to conflicting data on virologic efficacy and long-term outcomes. This meta-analysis investigated studies reporting the sustained virologic response (SVR), recurrence, and overall survival in patients with HCC treated with DAAs. Eighty-eight studies were included, comprising 8,839 patients with HCC who were treated with DAAs. The pooled SVR rate in patients with HCC was 89% (95% confidence interval [CI], 87% to 91%). However, the SVR varied significantly depending on tumor viability; patients with non-viable HCC had the highest SVR (91%), followed by those with mixed (88%) and viable HCC (84%). The SVR was significantly lower in the HCC group than in the non-HCC group (risk ratio [RR], 0.95; 95% CI, 0.92 to 0.98). In 28 studies reporting on recurrence outcomes, patients with HCC who were treated with DAAs had a 40% lower recurrence risk than non-DAA-treated patients (RR, 0.60; 95% CI, 0.49 to 0.75). Pooled analyses of adjusted estimates showed that DAA treatment was independently associated with reduced recurrence (hazard ratio [HR], 0.47; 95% CI, 0.32 to 0.70) and all-cause mortality (HR, 0.40; 95% CI, 0.31 to 0.51). An SVR was linked to an improved prognosis, with lower recurrence (HR, 0.43; 95% CI, 0.31 to 0.61) and mortality (HR, 0.41; 95% CI, 0.14 to 1.24) risks. DAAs are effective in patients with HCC. Despite slightly lower SVR rates than in patients without HCC, the overall benefits support the use of antiviral therapy in patients with HCC.

Background: Hepatocellular carcinoma (HCC) remains a major cause of cancer mortality worldwide, yet reliable molecular biomarkers for early diagnosis and prognosis are limited. The cytochrome P450 (CYP) enzyme superfamily plays central roles in hepatic metabolism and tumor biology, but its global dysregulation in HCC has not been comprehensively defined. Methods: Here, we systematically analyzed 57 CYP genes across multi-cohort transcriptomic datasets (GepLiver, TCGA-LIHC, HCCDB2.0) to delineate their diagnostic, prognostic, and functional significance. Results: CYP2R1 emerged as a consistently upregulated and clinically relevant member, showing excellent diagnostic accuracy (AUC = 0.95, 95% CI: 0.94–0.98, p < 0.001) and strong overexpression validated across independent cohorts and spatial transcriptomics. Prognostic modeling identified CYP2C9 (favorable) and CYP26B1 (unfavorable) as independent survival markers. Functional enrichment analyses revealed that high CYP2R1 expression was associated with activation of DNA repair and replication pathways (NES = 1.31, adjusted p = 1.05 × 10⁻⁵) and with co-expression of core repair genes such as POLA2, LIG1, and PCNA. Moreover, CYP2R1 correlated with myeloid-derived suppressor cell infiltration, suggesting an immunosuppressive phenotype. Conclusions: These findings establish CYP2R1 as a novel metabolic and immunogenomic biomarker in HCC, linking hepatic metabolism, genomic maintenance, and tumor immune modulation.