Risk of Bleeding in Hepatocellular Carcinoma Patients Treated with Atezolizumab/Bevacizumab: A Systematic Review and Meta-Analysis.

Introduction: The combination of atezolizumab/bevacizumab has emerged as an effective first-line treatment for advanced hepatocellular carcinoma (HCC). However, this therapy is potentially associated with bleeding complications, warranting a comprehensive analysis of their incidence and severity. This meta-analysis aims to synthesize available evidence from clinical trials and observational studies to quantify the prevalence of bleeding following atezolizumab/bevacizumab administration.

Methods: This meta-analysis focused on HCC treatment using atezolizumab/bevacizumab, particularly examining bleeding complications. It determined the prevalence of bleeding post-administration and compared the risk ratio with tyrosine kinase inhibitors (sorafenib or lenvatinib). Risk factors for bleeding complications were also evaluated.

Results: From 28 studies involving 3,895 patients, the pooled prevalence of bleeding side effects was 8.42% (95% CI: 5.72-11.54). Grade III or IV bleeding occurred in 4.42% (95% CI: 2.64-6.10) of patients, with grade V bleeding observed in 2.06% (95% CI: 0.56-4.22). Gastrointestinal bleeding, predominantly variceal, was the most common, with a prevalence of 5.48% (95% CI: 3.98-7.17). Subgroup analysis indicated variability in bleeding rates based on study design and geographical location. Atezolizumab/bevacizumab treatment exhibited a 2.11 times higher prevalence of bleeding compared to tyrosine kinase inhibitors (95% CI: 1.21-3.66). Meta-regression identified high body mass index (BMI) and higher proportion of albumin-bilirubin (ALBI) grade 3 as significant risk factors for bleeding complications.

Conclusion: Atezolizumab/bevacizumab therapy for advanced HCC carries a heightened risk of gastrointestinal bleeding, exceeding that of tyrosine kinase inhibitors. High BMI and higher ALBI grade are key predictors of bleeding complications, emphasizing the need for cautious patient selection and monitoring.