Taeang Arai

Background and Aims: For unresectable hepatocellular carcinoma (HCC), atezolizumab plus bevacizumab (Atez/Bev) and durvalumab plus tremelimumab (Dur/Tre) are widely used first‐line immunotherapies; however, biomarkers to guide regimen selection before treatment initiation remain undefined. We investigated whether baseline des‐γ‐carboxy prothrombin (DCP) modifies objective response rate (ORR) according to treatment regimen. Methods: We analysed 1464 patients with unresectable HCC treated with first‐line systemic therapy between 2020 and 2025 across multiple centers. Patients were divided into training and validation cohorts. Restricted cubic spline‐based interaction models were used to evaluate the association between baseline DCP and ORR according to treatment regimen. The predicted absolute difference in ORR (ΔORR) between regimens was examined across continuous DCP values, with a prespecified 10% threshold considered clinically meaningful. Robustness was assessed using leave‐one‐center‐out and repeated random‐split sensitivity analyses. Results: Overall ORR did not differ significantly between Atez/Bev and Dur/Tre. However, a significant interaction between baseline DCP and treatment regimen was observed in both cohorts. Atez/Bev showed a relatively stable ORR across DCP concentrations, whereas Dur/Tre showed a DCP‐dependent increase in ORR. At low DCP levels, Atez/Bev was favoured; with increasing DCP levels, the relative benefit shifted toward Dur/Tre. Sensitivity analyses confirmed a consistent DCP‐dependent transition pattern, although the DCP level at which ΔORR exceeded 10% varied across analyses, supporting a transition zone rather than a fixed cutoff. Conclusions: Baseline DCP may modify the relative treatment benefit between Atez/Bev and Dur/Tre in unresectable HCC. Rather than defining a rigid cutoff, DCP appears to delineate a continuum in which higher levels increasingly favour Dur/Tre, providing a practical framework for individualized immunotherapy selection.

Background and Aim: Atezolizumab plus bevacizumab (Atez/Bev) and durvalumab plus tremelimumab (Dur/Tre) are widely used first‐line immune checkpoint inhibitor (ICI)‐based regimens for unresectable hepatocellular carcinoma (uHCC). However, reliable biomarkers to inform treatment choice remain limited. This study evaluated whether the simplified BALAD‐based score (BALAD‐S) is associated with differential treatment outcomes. Materials and Methods: This multicenter retrospective study included 752 Japanese patients with uHCC who received first‐line Atez/Bev ( n = 645) or Dur/Tre ( n = 107) between October 2020 and March 2026. BALAD‐S was calculated using bilirubin, albumin, AFP, AFP‐L3, and des‐γ‐carboxy prothrombin, and patients were stratified into BALAD‐S low (≤ 2) and high (≥ 3) groups. Treatment response, progression‐free survival (PFS), and overall survival (OS) were evaluated. Results: In the BALAD‐S low group, Atez/Bev showed longer PFS than Dur/Tre (10.5 vs. 4.3 months; p < 0.001), with longer OS also observed (28.3 vs. 19.4 months; p = 0.049). In the BALAD‐S high group, PFS did not differ significantly (7.0 vs. 5.3 months with Dur/Tre and Atez/Bev, respectively, p = 0.210), whereas OS was longer with Dur/Tre than with Atez/Bev (27.3 vs. 16.4 months; p = 0.022). Treatment effects differed by BALAD‐S category (HR for Dur/Tre vs. Atez/Bev: 1.50 in BALAD‐S low and 0.55 in BALAD‐S high; p for interaction = 0.005). In patients treated from 2023 onward, similar treatment‐effect heterogeneity was observed (HR 1.61 in BALAD‐S low and 0.59 in BALAD‐S high; p for interaction = 0.007). Conclusion: Outcomes with Atez/Bev and Dur/Tre differed according to BALAD‐S score. BALAD‐S may help identify patients with differential outcomes between first‐line ICI‐based regimens and represents a candidate stratification marker requiring validation.

Background and Aim: This study evaluated clinical outcomes in patients with hepatocellular carcinoma (HCC) treated with durvalumab plus tremelimumab (Dur/Tre) in routine practice, stratified by whether they fulfilled the eligibility criteria of the phase 3 HIMALAYA trial. Methods: A total of 412 patients with unresectable HCC receiving Dur/Tre at 30 Japanese institutions were enrolled. Of these, 92 fulfilled the HIMALAYA trial eligibility criteria (HIMALAYA group) and 320 did not (non‐HIMALAYA group). Results: Median progression‐free survival (PFS) was 5.4 months in the HIMALAYA group and 3.0 months in the non‐HIMALAYA group ( p = 0.012). Multivariable analysis identified body mass index ≥ 25 kg/m ² (hazard ratio [HR], 0.780; 95% confidence interval [CI], 0.612–0.993; p = 0.044) and portal vein invasion (HR, 1.509; 95% CI, 1.049–2.170; p = 0.027) as independent predictors of PFS. Median overall survival (OS) was 19.4 months in the HIMALAYA group versus 15.4 months in the non‐HIMALAYA group ( p = 0.014). Multivariable analysis showed Eastern Cooperative Oncology Group performance status ≥ 1 (HR, 1.878; 95% CI, 1.253–2.814; p = 0.002) and albumin–bilirubin grade ≥ 2 (HR, 2.032; 95% CI, 1.319–3.318; p = 0.001) as independent determinants of OS. Any‐grade endocrine dysfunction occurred in 13 (14.1%) and 21 (6.6%) patients ( p = 0.030), with grade ≥ 3 events in 5 (5.4%) and 2 (0.6%), respectively ( p = 0.007). Subgroup analysis suggested that non‐HIMALAYA patients with ALBI grade 1 may have OS comparable to the HIMALAYA group. Conclusions: Patients fulfilling the HIMALAYA trial eligibility criteria and those who did not but had preserved hepatic function demonstrated favorable survival outcomes with Dur/Tre.

Background and Aims: Entecavir and tenofovir alafenamide fumarate are potent nucleos(t)ide analogues used to treat chronic hepatitis B; however, their differential effects on hepatitis B surface antigen expression decline remain unclear. In this multicentre study, we aimed to examine the hepatitis B surface antigen‐reducing effects. Methods: Treatment‐naïve chronic hepatitis B patients who received entecavir ( n = 491) or tenofovir alafenamide fumarate ( n = 463) were enrolled. After 1:1 propensity score matching for sex, baseline hepatitis B surface antigen, hepatitis B e‐antigen status, hepatitis B virus deoxyribonucleic acid, fibrosis‐4 score, and the presence of hepatocellular carcinoma, data from 334 patients per group were analysed. Linear mixed‐effects models were used to assess longitudinal changes over 4 years. Results: Tenofovir alafenamide fumarate was associated with a greater decline in hepatitis B surface antigen compared to entecavir, with significant differences at 1, 3, and 4 years. Its superiority was most pronounced in patients aged < 65 years, those without cirrhosis, those with baseline hepatitis B surface antigen ≥ 3.0 log IU/mL, those with either negative or positive hepatitis B e‐antigen, and those with baseline hepatitis B virus deoxyribonucleic acid < 4.4 log IU/mL. Both treatments achieved comparable virological suppression and alanine aminotransferase normalisation. The estimated glomerular filtration rate showed a transient decline with tenofovir alafenamide fumarate in year 1, which subsequently diminished and may partly reflect the inhibition of tubular creatinine secretion. Conclusions: This evidence supports personalised treatment selection for functional cure. Trial Registration: The UMIN Clinical Trials Registry 000034362

Aim: Atezolizumab plus bevacizumab (Atez/Bev) in unresectable hepatocellular carcinoma (uHCC) is a standard first‐line therapy, but bleeding‐related adverse events (BL‐AEs) remain clinically concerning, particularly in real‐world patients with advanced tumors, portal hypertension, or undergoing antithrombotic therapy (ATT). We aimed to evaluate whether ATT increases bleeding risk, identify pretreatment factors associated with BL‐AEs, and assess the impact of bleeding on progression‐free survival (PFS) and overall survival (OS) in patients with uHCC treated with Atez/Bev. Methods: This multicenter retrospective study included 981 patients with uHCC treated with Atez/Bev. BL‐AEs were identified by clinical assessment, imaging, or endoscopy. Predictors of BL‐AEs were analyzed using logistic regression. Survival outcomes were compared by Kaplan–Meier method, with inverse probability weighting (IPW) applied to adjust for baseline imbalances. Results: BL‐AEs occurred in 102 patients (10.4%) and led to treatment discontinuation in 2.7% patients. Independent predictors of BL‐AEs included advanced tumor burden components, such as Vp ≥ 3 portal vein invasion and extrahepatic metastasis. Neither ATT nor portal hypertension–related factors were associated with bleeding risk. PFS did not differ between groups, whereas OS was significantly shorter in the BL group before IPW adjustment than after IPW adjustment. Conclusions: During Atez/Bev therapy, BL‐AEs were primarily associated with advanced tumor burden rather than with ATT or baseline portal hypertension–related factors. Although BL‐AEs were associated with poorer OS, this should be interpreted cautiously, as bleeding may reflect aggressive tumor biology rather than a direct mortality cause. Careful monitoring for bleeding complications may be required in patients with advanced tumor burden receiving Atez/Bev.

Background & Aims: Systemic therapy is the standard care for patients with hepatocellular carcinoma (HCC) and extrahepatic metastases. However, the relative prognostic importance of the intrahepatic tumor burden versus the extent of extrahepatic disease in the era of immune checkpoint inhibitor‐based therapy remains unclear. This study aimed to clarify the prognostic impact of intrahepatic tumor burden in patients with metastatic HCC treated with atezolizumab plus bevacizumab (Atez/Bev). Methods: We conducted a multicenter retrospective study of patients with HCC and extrahepatic metastases who received first‐line Atez/Bev therapy. The intrahepatic tumor burden was assessed using established radiologic parameters, including tumor size, number, and vascular invasion, and the patients were stratified according to the intrahepatic disease severity. Overall survival (OS) was defined as the primary endpoint. Survival analyses were carried out using the Kaplan–Meier method and Cox proportional hazards models. Results: Overall, 306 patients were included in the study. During a median follow‐up of 13.5 months, the OS differed markedly according to the intrahepatic tumor burden and macrovascular invasion. The median OS was 17.8 months in the mild MVI group, 12.2 months in the severe MVI group, 29.8 months in the UT7 in group, and 15.2 months in the UT7 out group ( p < 0.001). In multivariate analysis, the intrahepatic tumor burden remained an independent predictor of OS, whereas the extent and number of extrahepatic metastatic sites were not significantly associated with prognosis. Conclusions: In metastatic HCC treated with atezolizumab plus bevacizumab, intrahepatic tumor burden was associated with overall survival and may be useful for risk stratification and treatment optimization.

Durvalumab plus tremelimumab (Dur/Tre) is the first‐line treatment for unresectable hepatocellular carcinoma (uHCC). Immune‐mediated adverse events (imAEs) are common; however, the impact of specific imAE types on treatment persistence and outcomes remains unclear. This study evaluated the clinical features and prognostic implications of common imAEs during Dur/Tre therapy. This multicenter retrospective analysis included 351 patients with uHCC treated with Dur/Tre (January 2023 and February 2025), of whom 69 developed imAEs and had a follow up period of > 1 month. Baseline characteristics, period to onset and severity of imAEs, corticosteroid use, treatment continuation and subsequent therapies, progression‐free survival (PFS) and overall survival (OS) were analyzed according to imAE types. Among 69 patients, 34 developed colitis, 19 developed hepatitis and 16 developed endocrine disorders. The median age was 72 years, with 13% classified as Child‐Pugh B. The median period to imAE onset was 0.75 months (interquartile range, 0.38–1.25). Steroids at a dose equivalent to > 20 mg of prednisolone were administered to 46 patients (66.7%). Two‐thirds of the patients continued Dur/Tre despite imAE onset. Colitis occurred the earliest (median 0.54 months) and had the highest steroid requirement (82.4%), frequently resulting in treatment interruption. Endocrine disorders showed the highest continuation rate (75%). Although not statistically significant in PFS (median 11.2 months: 95% CI, 1.96–not achieved (NA), p = 0.450) and OS (95% CI, NA–NA, p = 0.134), compared with colitis (PFS; 4.3 months, OS; not reached) and hepatitis (PFS; 3.4 months, OS: 14.6 months). These findings should be interpreted cautiously given the small sample size. Subsequent systemic or locoregional therapies, including tyrosine kinase inhibitors, atezolizumab plus bevacizumab and interventional therapies, were administered to more than 65% of the patients. In Dur/Tre therapy, imAE impact varied by type; endocrine disorders are linked to better treatment continuation and prognosis, whereas colitis and hepatitis often lead to treatment interruption. Trial Registration: Takasaki General Medical Center (IRB No. TGMC2024–03)

Aim: To assess whether non‐invasive test cutoffs recommended in Western guidelines appropriately identify treatment‐eligible fibrosis (F2–3) in Japanese patients with MASLD. Methods: We analyzed Japanese patients with biopsy‐proven MASLD who underwent FIB‐4 screening, followed by VCTE or ELF testing. Patients with FIB‐4 < 1.3 at age < 65 years or < 2.0 at age ≥ 65 years were excluded, leaving 838 patients in the VCTE cohort and 657 in the ELF cohort. Receiver operating characteristic (ROC) analyses were used to derive cohort‐specific cutoffs. Results: For predicting F2–4, a VCTE cutoff of 8 kPa yielded sensitivity/specificity/positive predictive value (PPV)/negative predictive value (NPV) of 83.7%/61.2%/80.2%/66.7%. For predicting F4, a VCTE cutoff of 15 kPa yielded 88.1%/76.8%/22.3%/98.8%. For ELF, a cutoff of 9.2 for predicting F2–4 yielded 94.7%/23.1%/66.5%/73.1%. For predicting F4, a ELF cutoff of 10.5 yielded 80.0%/58.4%/9.9%/98.1%. Using ROC‐derived cutoffs (VCTE 9.5 and 15 kPa; ELF 10.3 and 10.8), the performance for predicting F2–4 was 74.0%/75.6%/85.1%/60.8% for VCTE and 66.2%/73.9%/80.6%/57.1% for ELF, whereas the performance for predicting F4 was 88.1%/76.8%/22.3%/98.8% for VCTE and 77.1%/69.8%/12.6%/98.2% for ELF. Among test‐positive patients, the distribution of F0–1/F2–3/F4 was 28.7%/69.2%/2.1% for VCTE 8–15 kPa and 47.8%/49.8%/2.4% for ELF 9.2–10.5. Conclusions: Western guideline‐recommended VCTE and ELF cutoffs showed high sensitivity in Japanese patients and can be used to exclude cirrhosis. Because the likelihood of cirrhosis increases with VCTE > 15 kPa or ELF > 10.5, treatment eligibility should be determined in conjunction with other clinical information, and confirmatory assessment should be considered when appropriate.

Background and Aims: This multicenter retrospective study in Japan aimed to investigate the prognostic significance of lymphocyte‐to‐monocyte ratio (LMR) in patients with unresectable hepatocellular carcinoma (HCC) treated with durvalumab plus tremelimumab (Dur/Tre). Methods: A total of 377 patients with HCC and treated with Dur/Tre across 30 institutions in Japan were included in this multicenter study. Time‐dependent receiver operating characteristic (ROC) analysis was performed to determine the optimal LMR cut‐off value. Hazard ratio (HR) spline curve analysis was used to identify the optimal LMR range for predicting progression‐free survival (PFS) and overall survival (OS). Results: Time‐dependent ROC analysis identified an optimal LMR cut‐off value of 2.52 for predicting median OS. Multivariate analysis demonstrated that an LMR of ≥ 2.52 was independently associated with superior PFS (HR: 0.777) and OS (HR: 0.657). The median PFS was 2.6 months in patients with an LMR of < 2.52, compared with 3.5 months in those with an LMR of ≥ 2.52 ( p = 0.022). The median OS was 12.8 months in patients with an LMR of < 2.52, compared with 23.4 months in those with an LMR of ≥ 2.52 ( p < 0.001). The disease control rate was significantly higher in the high LMR group ( p = 0.032). The HR spline curve analysis revealed that an LMR range of approximately 1.8–2.6 represents an optimal cut‐off for predicting both PFS and OS. Conclusions: LMR is a readily accessible prognostic biomarker for both PFS and OS in patients with unresectable HCC treated with Dur/Tre, and may serve as a practical tool for risk stratification in clinical practice.