Abstract
BACKGROUND/AIMS: Although atezolizumab plus bevacizumab has significantly improved the life expectancy of patients with unresectable hepatocellular carcinoma (HCC), it also increases bleeding risks. This study aimed to identify factors associated with bleeding events and evaluate their impact on prognosis. METHODS: Patients treated with atezolizumab plus bevacizumab as first-line therapy for unresectable HCC were retrospectively reviewed. Patients with high-risk varices were treated before therapy initiation. The primary endpoint was the incidence of bleeding events and secondary endpoints were overall survival (OS) and disease control rate (DCR). RESULTS: Among 123 patients, 81 had varices detected via esophagogastroduodenoscopy or computed tomography (varices group) while 42 did not (non-varices group). During a median follow-up of 11.1 months, bleeding events occurred in 15 patients, with 14 of occurring in the varices group. The cumulative incidence of bleeding in the varices group was 7.7%, 21.3%, or 32.6% at 6, 12, or 18 months, respectively, significantly higher than that (0.0%) in the non-varices group (p=0.001). No significant difference in OS was observed between the groups after inverse probability of treatment weighting (hazard ratio [HR], 0.84; 95% confidence interval [CI], 0.49 to 1.46; p=0.54). Bleeding events were not significantly associated with OS after inverse probability of treatment weighting (HR, 0.68; 95% CI, 0.35 to 1.33; p=0.26). However, the DCR was significantly higher in the varices group than in the non-varices group (80.2% vs 54.8%; p=0.006). CONCLUSIONS: In unresectable HCC patients treated with atezolizumab plus bevacizumab, varices increase bleeding risk. However, proactive management and careful monitoring could mitigate their impact on OS and help increase the DCR.