Supplementary Material for: Analysis of Immune-Related Adverse Events of Atezolizumab and Bevacizumab in Patients with Hepatocellular Carcinoma: A Multicenter Cohort Study

Background: Despite the emergence of atezolizumab and bevacizumab (A+B) as standard first-line systemic therapy for unresectable hepatocellular carcinoma (HCC), a comprehensive understanding of the clinical significance of immune-related adverse events (irAEs) remains limited. We aimed to assess the impact of irAEs on patients with HCC undergoing A+B treatment.
Methods: This multicentre retrospective study included consecutive patients with HCC who were treated with the A+B regimen from September 2020 to December 2022. Patients were categorised into three groups based on the severity of irAEs, ranging from those without any experience of irAEs to those with severe irAEs, classified as grade 3 or higher.
Results: This study included 150 patients with HCC, with a mean age of 63.3 years. Among them, 93.3% of patients were classified as Barcelona Clinic Liver Cancer stage C, 52.0% had portal vein tumour thrombosis (PVTT), and 60.7% extrahepatic spread. Patients were classified as follows: Group 1 (n = 84) had no irAEs, Group 2 (n = 37) had mild irAEs (grade 1–2), and Group 3 (n = 29) had severe irAEs (grade ≥ 3). The median overall survival (OS), progression-free survival (PFS), and time-to-treatment discontinuation (TTD) were 13.6, 5.7, and 3.6 months, respectively. Group 2 demonstrated significantly superior OS compared to Group 1 (9.5 months) and Group 3 (5.6 months), with a median OS of 23.0 months (p < 0.001). Furthermore, Group 2 demonstrated significantly better outcomes in terms of PFS and TTD compared to both Group 1 and Group 3 (p < 0.001 for both). Multivariate analysis identified mild irAEs (hazard ratio [HR], 0.353; p = 0.010), ALBI grade 1 (HR, 0.389; p = 0.006), Child-Pugh class A (HR, 0.338; p = 0.002), and the absence of PVTT (HR, 0.556; p = 0.043) as independent predictors of better OS.
Conclusions: Our study highlights the significant impact of irAE severity on the outcomes of patients with HCC receiving A+B. Notably, the occurrence of mild irAEs was independently associated with favourable survival, suggesting their potential role as surrogate indicators of HCC prognosis.

背景：尽管阿替利珠单抗联合贝伐珠单抗（A+B）已成为不可切除肝细胞癌（hepatocellular carcinoma, HCC）的标准一线全身治疗方案，但目前对免疫相关不良事件（immune-related adverse events, irAEs）的临床意义仍缺乏全面认知。本研究旨在评估免疫相关不良事件对接受A+B方案治疗的肝细胞癌患者的影响。
方法：本项多中心回顾性研究纳入了2020年9月至2022年12月期间接受A+B方案治疗的连续入组肝细胞癌患者。根据免疫相关不良事件的严重程度将患者分为三组：无免疫相关不良事件组、轻度免疫相关不良事件组（1~2级）及重度免疫相关不良事件组（3级及以上）。
结果：本研究共纳入150例肝细胞癌患者，平均年龄为63.3岁。其中93.3%的患者为巴塞罗那临床肝癌分期C期，52.0%合并门静脉癌栓（portal vein tumour thrombosis, PVTT），60.7%存在肝外转移。患者分组情况如下：第1组（n=84）无免疫相关不良事件，第2组（n=37）为轻度免疫相关不良事件（1~2级），第3组（n=29）为重度免疫相关不良事件（≥3级）。患者的中位总生存期（overall survival, OS）、无进展生存期（progression-free survival, PFS）及治疗终止时间（time-to-treatment discontinuation, TTD）分别为13.6个月、5.7个月及3.6个月。第2组的中位总生存期为23.0个月，显著优于第1组（9.5个月）与第3组（5.6个月）（p<0.001）。此外，第2组在无进展生存期与治疗终止时间方面的结局也显著优于第1组与第3组（两者p均<0.001）。多因素分析显示，轻度免疫相关不良事件（风险比[hazard ratio, HR]=0.353；p=0.010）、ALBI分级1级（HR=0.389；p=0.006）、Child-Pugh分级A级（HR=0.338；p=0.002）及无门静脉癌栓（HR=0.556；p=0.043）是总生存期更优的独立预测因素。
结论：本研究证实，免疫相关不良事件的严重程度对接受A+B方案治疗的肝细胞癌患者的临床结局具有显著影响。值得注意的是，轻度免疫相关不良事件的发生与良好的生存结局独立相关，提示其可作为肝细胞癌预后的替代标志物。