AASLD 2025 | 晚期肝细胞癌治疗新进展：STRIDE、TKIs疗效揭秘

近日，全球肝病领域顶级学术盛会-美国肝病研究学会(AASLD)年度大会 The Liver Meeting 2025 在美国•华盛顿会议中心正式召开。本次会议将汇聚肝病研究领域的最新科研硕果与前沿探索方向，为全球参会者精心构筑一个深化肝病理论认知、拓展专业视野边界的学术交流殿堂。

为助力广大读者精准捕捉会议的学术精髓，及时洞悉领域前沿动态，肝胆相照平台特别甄选会议摘要中的热点研究内容，本篇专题报道聚焦“肝细胞癌”这一关键领域，为广大同仁提供最新研究进展，展望其在临床实践中的广阔应用前景。

热点研究汇总

一

3169

晚期/不可切除肝细胞癌患者接受 STRIDE 与 阿替利珠单抗-贝伐珠单抗治疗的临床结局

CLINICAL OUTCOMES IN PATIENTS WITH ADVANCED/UNRESECTABLE HEPATOCELLULAR CARCINOMA TREATED WITH STRIDE VERSUS ATEZOLIZUMAB-BEVACIZUMAB

作者：Michael Li¹, Huat Lim¹, John Gordan¹, Jocelin Chen¹, Quincy Harris¹, Jennifer Lai¹, Lawrence Fong², Robin Kelley¹

¹University of California, San Francisco；²Fred Hutchinson Cancer Center

背景

单次特瑞普利单抗联合定期度伐利尤单抗（STRIDE）方案与阿替利珠单抗-贝伐珠单抗（atezo/bev）方案均为晚期肝细胞癌（HCC）的一线系统治疗，可改善患者生存，但两者尚未直接比较。本研究旨在评估不同方案对临床结局的影响。

Background：Single tremelimumab regular interval durvalumab (STRIDE) and atezolizumab-bevacizumab (atezo/bev) are first-line systemic therapy options which improve survival in advanced hepatocellular carcinoma (HCC) but have not been directly compared. We sought to evaluate the impact of regimen on outcomes.

方法

我们开展了一项单中心回顾性队列研究，纳入接受 STRIDE 或 atezo/bev 治疗的 HCC 患者。主要终点为总生存期（OS）和真实世界无进展生存期（rwPFS，定义为临床评估的PFS）；次要终点为治疗持续时间（TOT），作为PFS的替代指标。为减少适应证混杂影响，研究建立了预测STRIDE治疗（相对atezo/bev）的倾向评分模型，并通过逆概率加权（IPTW）法进行平衡后Cox回归分析。灵敏度分析采用调整相同混杂因素的标准Cox回归。

Methods：We performed a single-center retrospective cohort study of HCC patients who received STRIDE or atezo/bev. Primary outcomes were overall survival (OS) and real-world progression-free survival (rwPFS), defined as clinically-assessed PFS. The secondary outcome was time on treatment (TOT) as a surrogate measure of PFS. To account for confounding by indication, a propensity score model predicting STRIDE treatment (vs atezo/bev) was developed and then used in inverse probability of treatment weighting (IPTW) to create balanced groups for Cox regression. Sensitivity analyses were performed using standard Cox regression adjusting for the same confounders included in the propensity score.

 结果

共纳入203例患者，其中76例（37%）接受STRIDE治疗，127例（63%）接受atezo/bev治疗。两组在肝功能、肿瘤分期、病因及出血史方面差异无统计学意义。STRIDE组接受一线系统治疗的比例较低（67% vs 89%，P<0.001），但既往肝外血栓栓塞病史更多（12% vs 3%，P=0.014）。

与atezo/bev组相比，STRIDE组的疾病控制率较低（56% vs 72%，P=0.051）。IPTW平衡后，两组的主要基线特征（包括治疗线次、血栓史）均匹配。IPTW-Cox回归分析显示，两组OS（HR 0.7，P=0.147）和rwPFS（HR 0.9，P=0.662）差异均无统计学意义。

然而，在获得疾病控制的患者亚组（n=122，占60%）中，STRIDE组表现出更长的OS（HR 0.5，95% CI 0.2–0.9，P=0.034）、rwPFS（HR 0.4，95% CI 0.2–0.9，P=0.018）和TOT（HR 0.4，95% CI 0.2–0.8，P=0.013）。

在该亚组的灵敏度分析中，标准Cox回归结果一致：STRIDE仍显著延长OS（aHR 0.4，95% CI 0.2–0.8，P=0.011）、rwPFS（aHR 0.3，95% CI 0.2–0.7，P=0.004）及TOT（aHR 0.3，95% CI 0.2–0.7，P=0.002）。在一线治疗且获得疾病控制的患者中，未调整的OS（中位数未达到 vs 1.8年）和rwPFS（中位数未达到 vs 0.8年）均显示STRIDE更优。

Results：Of 203 patients, 76 (37%) received STRIDE and 127 (63%) received atezo/bev. There were no differences in liver function, HCC stage, liver disease etiology, or bleeding history between the two groups. STRIDE patients were less likely to be receiving first-line systemic therapy (67% vs 89%, p<0.001) and more likely to have prior extrahepatic thromboembolism (12% vs 3%, p=0.014). STRIDE was associated with a lower disease control rate (56% vs 72%, p=0.051) compared to atezo/bev. After IPTW, key baseline characteristics were balanced, including line of therapy and prior thromboembolism (Love plot, Fig A). On IPTW-Cox regression, there were no differences in OS (HR 0.7, p=0.147) or rwPFS (HR 0.9, p=0.662) when comparing STRIDE vs atezo/bev.

However, in the subgroup of patients who achieved disease control (n=122, 60%), STRIDE was associated with longer OS (HR 0.5, 95% CI 0.2-0.9, p=0.034), rwPFS (HR 0.4, 95% CI 0.2-0.9, p=0.018), and TOT (HR 0.4, 95% CI 0.2-0.8, p=0.013) compared to atezo/bev. Sensitivity analyses in the disease control subgroup using standard Cox regression resulted in similar hazard ratios, with STRIDE remaining associated with longer OS (aHR 0.4, 95% CI 0.2-0.8, p=0.011), rwPFS (aHR 0.3, 95% CI 0.2-0.7, p=0.004), and TOT (aHR 0.3, 95% CI 0.2-0.7, p=0.002). In the first-line subgroup of patients with disease control, unadjusted OS (median not reached vs 1.8 years, Fig B) and rwPFS (median not reached vs 0.8 years; Fig C) favored STRIDE.

结论

与atezo/bev相比，STRIDE的疾病控制率较低，这可能与atezo/bev方案中的抗血管生成成分（贝伐珠单抗）产生反应有关。然而，在达到疾病控制的患者中，STRIDE治疗显著延长了OS、rwPFS和TOT。本研究结果支持双免疫检查点抑制可能带来更持久的免疫应答。

Conclusion：STRIDE may be associated with a lower disease control rate compared to atezo/bev, which could reflect that some atezo/bev-treated patients respond to the antiangiogenic component (bev). However, in patients who do achieve disease control, STRIDE treatment was associated with longer OS, rwPFS, and TOT. Our findings support the hypothesis that dual checkpoint inhibition may be associated with more durable immune response.

二

3243

晚期肝细胞癌在阿替利珠单抗/贝伐珠单抗治疗后使用酪氨酸激酶抑制剂（TKIs）的疗效：系统综述与Meta分析

UTILITY OF TYROSINE KINASE INHIBITORS AFTER ATEZOLIZUMAB/BEVACIZUMAB IN ADVANCED HEPATOCELLULAR CARCINOMA: A SYSTEMATIC REVIEW AND META-ANALYSIS

作者：Ronald Blanco Montecino¹、Nehemias Guevara Rodriguez²、Iqra Arshad³、Daniel Mendoza Quispe¹、Noemy Coreas⁴、Yassine Kilani³、Syeda Fatima Kamal³、Kamran Qureshi⁵、John Richart²

¹Department of Internal Medicine, St Barnabas Hospital, Bronx, NY

²Division of Hematology-Oncology and BMT, Saint Louis University, St. Louis, MO

³Division of General Internal Medicine, Saint Louis University School of Medicine, St. Louis, MO

⁴National University of El Salvador, Social Security System, Division of OBGYN-Oncology

⁵Division of Gastroenterology and Hepatology, Saint Louis University School of Medicine, St. Louis, MO

背景

阿替利珠单抗联合贝伐珠单抗（Atezolizumab + Bevacizumab，简称Atezo/Bev）目前已确立为晚期肝细胞癌（HCC）患者的一线标准治疗方案之一。然而，在该方案进展后的治疗选择仍在不断发展中，尤其是不同二线酪氨酸激酶抑制剂（TKIs）的作用及相对疗效尚不明确。为明确最佳二线治疗方案，本研究对Atezo/Bev治疗失败后使用不同TKIs的疗效进行了系统综述与Meta分析。

Background：Atezolizumab, in combination with Bevacizumab, is currently established as one of the standard first-line regimens for patients with advanced hepatocellular carcinoma (HCC). However, the therapeutic landscape following the progression of this regimen is evolving, particularly regarding the role and relative efficacy of various second-line tyrosine kinase inhibitors (TKIs). To clarify the optimal second-line option, we conducted a systematic review and meta-analysis to evaluate the experience with various TKIs as second-line therapy in patients with advanced HCC who experienced disease progression on Atezolizumab/Bevacizumab.

方法

系统检索PubMed、Scopus、Embase和Cochrane临床试验数据库，纳入截至2024年10月前的所有相关研究。符合条件的研究需报告接受TKI二线治疗患者的中位总生存期（OS）和/或无进展生存期（PFS）。使用Excel工具（DECoMA）从中位数和四分位距估算平均值和标准差，并采用随机效应模型整合各研究的中位OS和PFS。统计分析使用Stata 18软件。

Methods：A systematic search of PubMed, Scopus, Embase, and the Cochrane Central Register of Controlled Trials was performed from inception to October 2024. Eligible studies reported median overall survival (OS) and/or progression-free survival (PFS) for patients receiving second-line TKIs. Mean and standard deviation estimates were derived from medians and interquartile ranges using an Excel-based tool for estimating sample mean and standard deviation (SD) (Data Estimation and Conversion for Meta-Analysis [DECoMA]). A random-effects model was applied to pool median OS and PFS across studies. Statistical analyses were conducted using Stata version 18.

 结果

共纳入9项回顾性研究，总计907例在Atezo/Bev治疗进展后的晚期HCC患者。所有患者均接受TKI二线治疗：Lenvatinib（n=442）、Sorafenib（n=365）、Regorafenib（n=65）和Cabozantinib（n=35）。总体合并中位OS为11.0个月（95% CI：7.99–14.00）。

分治疗组分析：Lenvatinib组为12.99个月（95% CI：7.76–18.23），Sorafenib组为10.28个月（95% CI：2.71–17.86），Cabozantinib组为8.79个月（95% CI：5.74–11.85），而Regorafenib数据有限。各药物之间的OS差异无统计学意义。合并PFS为3.46个月（95% CI：2.99–3.92）。其中，Lenvatinib的PFS最长，为3.95个月（95% CI：3.63–4.27）；Sorafenib为2.46个月（95% CI：2.07–2.85）；Cabozantinib和Regorafenib分别为3.41个月（95% CI：1.26–5.56）和3.27个月（95% CI：2.22–4.33），三者间差异亦无统计学意义。

Results： Nine retrospective studies were included, with 907 patients with advanced HCC who progressed on Atezolizumab/Bevacizumab. All patients received a TKI as second-line therapy: Lenvatinib (n = 442), Sorafenib (n =365), Regorafenib (n =65), or Cabozantinib (n = 35). Across all studies, overall survival (OS) was 11.0 months (95% CI: 7.99–14.00). When stratified by treatment, the median OS was 12.99 months with Lenvatinib (95% CI: 7.76–18.23), 10.28 months with Sorafenib (95% CI: 2.71–17.86), and 8.79 months with Cabozantinib (95% CI: 5.74–11.85), and data were limited for Regorafenib. No statistically significant differences in OS were observed between the agents. The pooled mean progression-free survival (PFS) was 3.46 months (95% CI: 2.99–3.92). Among the evaluated TKIs, Lenvatinib showed the longest PFS at 3.95 months (95% CI: 3.63–4.27). Sorafenib had a PFS of 2.46 months (95% CI: 2.07–2.85). Cabozantinib and Regorafenib demonstrated mean PFS of 3.41 months (95% CI: 1.26–5.56) and 3.27 months (95% CI: 2.22–4.33), respectively. No statistically significant differences in PFS were observed among Cabozantinib, Sorafenib, and Regorafenib.

结论

在Atezo/Bev治疗进展后的晚期HCC患者中，TKIs作为二线治疗方案仅显示出有限的临床获益。虽然不同药物间的总生存期差异无统计学意义，但Lenvatinib的无进展生存期（PFS）相对更长，而Cabozantinib与Regorafenib无显著差异。仍需进一步的大规模前瞻性研究来验证这些结果，并优化该治疗阶段的药物序贯策略。

Conclusion：Tyrosine kinase inhibitors showed modest clinical benefits as second-line options for advanced HCC after progression on Atezolizumab/Bevacizumab. While overall survival did not differ significantly among agents, Lenvatinib showed a longer PFS, with no significant differences seen with Cabozantinib or Regorafenib. Further large-scale prospective studies are needed to validate these findings and refine treatment sequencing in this setting.

三

2406

利福昔明可降低接受仑伐替尼治疗且伴脾肾分流的肝细胞癌患者新发肝性脑病的风险

TARGETING THE HIDDEN RISK: RIFAXIMIN LOWERS THE RISK OF NEW ONSET OF HEPATIC ENCEPHALOPATHY IN PATIENTS WITH HEPATOCELLULAR CARCINOMA AND SPLENORENAL SHUNT RECEIVING LENVATINIB

作者：Takumi Kawaguchi¹, Shigeo Shimose¹, Hideki Iwamoto², Takashi Niizeki¹, Tomotake Shirono¹, Etsuko Moriyama¹

¹Kurume University School of Medicine, ²Iwamoto Internal Medicine Clinic

背景

仑伐替尼可有效抑制肝细胞癌（HCC）的肿瘤血管生成及生长，从而改善患者预后。然而，肝性脑病（HE）是仑伐替尼治疗中需要停药的严重不良事件之一。利福昔明是一种口服、极少吸收的抗菌药物，可降低肝硬化患者肝性脑病复发的风险。本研究旨在探讨接受仑伐替尼治疗的HCC患者发生HE的风险因素，并评估利福昔明在预防HE及维持仑伐替尼剂量中的作用。

Background：Lenvatinib effectively inhibits tumor angiogenesis and growth in hepatocellular carcinoma (HCC), improving prognosis. However, hepatic encephalopathy (HE) is a serious adverse event of lenvatinib that often necessitates drug discontinuation. Rifaximin, a minimally absorbed oral antimicrobial, reduces the recurrence of HE in patients with liver cirrhosis. This study aimed to identify risk factors for HE in HCC patients receiving lenvatinib and to evaluate the usefulness of rifaximin for HE prevention and lenvatinib dosing.

方法

本研究为双中心回顾性观察性队列研究，共纳入216例接受仑伐替尼治疗的HCC患者。HE的诊断依据扑翼样震颤及/或从定向障碍至昏迷的精神症状。在开始仑伐替尼治疗前，血氨水平≥80 μg/dL的患者接受口服利福昔明（1200 mg/天）治疗。观察期中位数为480天（范围：1–3657天），评估HE的发生率。仑伐替尼剂量通过相对剂量强度（RDI）进行评估，并采用决策树算法分析与HE相关的患者特征。

Methods：This two-center retrospective observational cohort study enrolled 216 HCC patients treated with lenvatinib. HE was diagnosed based on the presence of asterixis and/or mental symptoms ranging from disorientation to coma. Before starting lenvatinib, patients with hyperammonemia (≥80 μg/dL) received oral rifaximin (1,200 mg/day). The incidence of HE was evaluated over a median observational period of 480 (range: 1–3,657) days. Lenvatinib dosage was assessed by relative dose intensity (RDI). A decision-tree algorithm was used to identify profiles associated with HE.

 结果

HE的总体发生率为8.8%。HE组与非HE组在年龄、性别、BMI及肿瘤分期方面无显著差异。尽管两组血氨水平无统计学差异，但HE组的脾肾分流发生率显著更高。多变量分析显示，脾肾分流是HE的独立危险因素（HR 16.4，95% CI：5.2–52.6，P<0.001）。决策树分析表明，在伴脾肾分流且ALBI 2级的患者中，未使用利福昔明者HE发生率为66.6%，而使用利福昔明者仅为16.6%。此外，RDI>70%的患者比例在未使用利福昔明组为20.0%，而在使用组为66.7%。

Results：The overall incidence of HE was 8.8%. There were no significant differences in age, sex, BMI, or tumor stage between the HE and non-HE groups. Although blood ammonia levels did not differ significantly, the prevalence of splenorenal shunt was markedly higher in the HE group. Multivariate analysis identified splenorenal shunt as an independent risk factor for HE (HR 16.4, 95% CI: 5.2–52.6, p<0.001). Decision-tree analysis revealed that among patients with splenorenal shunt and ALBI grade 2, HE occurred in 66.6% of those without rifaximin, compared to 16.6% among those receiving rifaximin. Furthermore, the proportion of patients with RDI >70% was 20.0% without rifaximin versus 66.7% with rifaximin.

结论

脾肾分流是接受仑伐替尼治疗的HCC患者发生HE的独立危险因素。利福昔明不仅可预防HE的发生，还可帮助患者维持足够的仑伐替尼剂量。对于伴脾肾分流及肝功能损害（ALBI 2级）的HCC患者，利福昔明可能是一种有益的辅助管理策略。

Conclusion： Splenorenal shunt is an independent risk factor for HE in HCC patients receiving lenvatinib. Rifaximin effectively prevented HE and helped maintain adequate lenvatinib dosing in patients with HCC, splenorenal shunt, and ALBI grade 2. These findings suggest that rifaximin may be beneficial for managing HCC patients treated with lenvatinib, particularly those with splenorenal shunt and hepatic dysfunction.